Handle Tag model for insensitive search

src/documents/admin.py

@@ -13,11 +13,13 @@
 from django.utils.safestring import mark_safe
 from djangoql.admin import DjangoQLSearchMixin
 
-from documents.actions import (add_tag_to_selected,
-                               remove_correspondent_from_selected,
-                               remove_tag_from_selected,
-                               set_correspondent_on_selected)
-from paperless.utils import slugify
+from documents.actions import (
+    add_tag_to_selected,
+    remove_correspondent_from_selected,
+    remove_tag_from_selected,
+    set_correspondent_on_selected
+)
+from paperless.utils import make_searchable
 
 from .models import Correspondent, Document, Log, Tag
 
@@ -176,7 +178,7 @@ class Media:
         "correspondent__name",
         "searchable_title",
         "searchable_content",
-        "tags__name",
+        "tags__searchable_name",
     )
     readonly_fields = ("added", "file_type", "storage_type",)
     list_display = ("title", "created", "added", "thumbnail", "correspondent",
@@ -343,7 +345,7 @@ def _html_tag(kind, inside=None, **kwargs):
         return format_html("<{} {}/>", kind, attributes)
 
     def get_search_results(self, request, queryset, search_term):
-        search_term = slugify(search_term)
+        search_term = make_searchable(search_term)
         return super().get_search_results(request, queryset, search_term)
 
 

src/documents/migrations/0023_document_searchable_content.py

@@ -3,7 +3,13 @@
 
 from django.db import migrations, models
 
-from paperless.utils import slugify as slugifyOCR
+
+def slugifyOCR(content):
+    return (
+        unicodedata.normalize("NFKD", content.casefold())
+        .encode("ASCII", "ignore")
+        .decode("utf-8")
+    )
 
 
 class Migration(migrations.Migration):
@@ -20,12 +26,16 @@ def casefold_forwards(apps, schema_editor):
                 doc.searchable_content = slugifyOCR(doc.content)
             doc.save()
 
-    def casefold_backwards(apps, schema_editor):
-        pass
+        Tag = apps.get_model("documents", "Tag")
+
+        for tag in Tag.objects.all():
+            tag.searchable_name = slugifyOCR(tag.name)
+            tag.save()
 
     def database_backwards(self, app_label, schema_editor, from_state, to_state):
         migrations.RemoveField(model_name="document", name="searchable_content"),
         migrations.RemoveField(model_name="document", name="searchable_title"),
+        migrations.RemoveField(model_name="tag", name="searchable_name"),
 
     operations = [
         migrations.AddField(
@@ -40,5 +50,12 @@ def database_backwards(self, app_label, schema_editor, from_state, to_state):
                 max_length=128, blank=True, db_index=True, editable=False
             ),
         ),
-        migrations.RunPython(casefold_forwards, casefold_backwards),
+        migrations.AddField(
+            model_name="tag",
+            name="searchable_name",
+            field=models.CharField(
+                max_length=128, blank=True, db_index=True, editable=False
+            ),
+        ),
+        migrations.RunPython(casefold_forwards, migrations.RunPython.noop),
     ]

src/documents/models.py

@@ -9,12 +9,11 @@
 import dateutil.parser
 from django.conf import settings
 from django.db import models
-from django.template.defaultfilters import slugify
 from django.utils import timezone
 from django.utils.text import slugify
 from fuzzywuzzy import fuzz
 
-from paperless.utils import slugify as slugifyOCR
+from paperless.utils import make_searchable
 
 from .managers import LogManager
 
@@ -185,6 +184,15 @@ class Tag(MatchingModel):
 
     colour = models.PositiveIntegerField(choices=COLOURS, default=1)
 
+    searchable_name = models.CharField(
+        max_length=128, blank=True, db_index=True, editable=False,
+    )
+
+    def save(self, *args, **kwargs):
+        if self.name is not None:
+            self.searchable_name = make_searchable(self.name)
+        return super().save(*args, **kwargs)
+
 
 class Document(models.Model):
 
@@ -283,9 +291,9 @@ def __str__(self):
 
     def save(self, *args, **kwargs):
         if self.title is not None:
-            self.searchable_title = slugifyOCR(self.title)
+            self.searchable_title = make_searchable(self.title)
         if self.content is not None:
-            self.searchable_content = slugifyOCR(self.content)
+            self.searchable_content = make_searchable(self.content)
         return super().save(*args, **kwargs)
 
     @property

src/documents/tests/test_tag_model.py

@@ -0,0 +1,12 @@
+from unittest import mock
+
+from django.test import TestCase
+
+from ..models import Tag
+
+
+class TestTag(TestCase):
+    def test_searchable_name(self):
+        tag = Tag.objects.create(name="Titlé")
+        self.assertEqual(tag.name, "Titlé")
+        self.assertEqual(tag.searchable_name, "title")

src/paperless/utils.py

@@ -1,7 +1,7 @@
 import unicodedata
 
 
-def slugify(content):
+def make_searchable(content):
     return (
         unicodedata.normalize("NFKD", content.casefold())
         .encode("ASCII", "ignore")