Merge pull request #250 from kmc-jp/fix-tag-name-collation

タグ名のひらがな・カタカナを区別して保存する

api/src/goduploader/model/tag.py

@@ -1,6 +1,6 @@
-from typing import List
 import unicodedata
 from datetime import datetime
+from typing import List
 from urllib.parse import urljoin
 
 from goduploader.config import app_config
@@ -25,12 +25,12 @@ class Tag(Base):
 
     # 正規化した (小文字に統一した) タグ名
     canonical_name = Column(
-        String(255),
+        String(255, collation='utf8mb4_general_ci'),
         nullable=False,
         unique=True,
     )
     # 表示されるタグ名
-    name = Column(String(255), nullable=False)
+    name = Column(String(255, collation='utf8mb4_general_ci'), nullable=False)
     artworks_count = Column(Integer, nullable=False, default=0)
 
     index_artworks_count = Index("tag_artworks_count", artworks_count)

api/src/migration/versions/4f308d092386_change_tag_name_collation_to_utf8mb4_.py

@@ -0,0 +1,30 @@
+"""change tag name collation to utf8mb4_general_ci
+
+Revision ID: 4f308d092386
+Revises: aa084f2c60f6
+Create Date: 2024-01-23 21:18:51.543256
+
+"""
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision = '4f308d092386'
+down_revision = 'aa084f2c60f6'
+branch_labels = None
+depends_on = None
+
+
+def upgrade():
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.alter_column('tag', 'canonical_name', type_=sa.types.VARCHAR(255, collation='utf8mb4_general_ci'))
+    op.alter_column('tag', 'name', type_=sa.types.VARCHAR(255, collation='utf8mb4_general_ci'))
+    # ### end Alembic commands ###
+
+
+def downgrade():
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.alter_column('tag', 'canonical_name', type_=sa.types.VARCHAR(255))
+    op.alter_column('tag', 'name', type_=sa.types.VARCHAR(255))
+    # ### end Alembic commands ###

api/tests/model/test_tag.py

@@ -30,6 +30,19 @@ def test_find_or_create():
     assert created_tags[1].name == "BBB"
     assert created_tags[1].canonical_name == "bbb"
 
+    # COLLATE=utf8mb4_0900_ai_ci でひらがなカタカナを区別せず、タグが作れていなかったことに対するリグレッションテスト
+    created_tags = Tag.find_or_create(["ロリ"])
+    session.commit()
+
+    assert created_tags[0].name == "ロリ"
+    assert created_tags[0].canonical_name == "ロリ"
+
+    created_tags = Tag.find_or_create(["ろり"])
+    session.commit()
+
+    assert created_tags[0].name == "ろり"
+    assert created_tags[0].canonical_name == "ろり"
+
 
 @pytest.mark.parametrize(
     "tag_name, expected",