Add automatic naming conventions for database relations

BUILD.md

@@ -20,7 +20,7 @@ poetry install
 When poetry is done installing all dependencies you can start using the tool.
 
 ```shell
-poetry run python -m uvicorn amt.main:app --log-level warning
+poetry run python -m uvicorn amt.server:app --log-level warning
 ```
 
 ### Suggested development ENVIRONMENT settings

amt/api/deps.py

@@ -44,7 +44,7 @@ def _create_env(
         directory: str | PathLike[typing.AnyStr] | typing.Sequence[str | PathLike[typing.AnyStr]],
         **env_options: typing.Any,  # noqa: ANN401
     ) -> Environment:
-        env: Environment = super()._create_env(directory, **env_options)  # pyright: ignore [reportUnknownMemberType, reportUnknownVariableType]
+        env: Environment = super()._create_env(directory, **env_options)  # pyright: ignore [reportUnknownMemberType, reportUnknownVariableType, reportArgumentType]
         env.add_extension("jinja2.ext.i18n")  # pyright: ignore [reportUnknownMemberType]
         return env  # pyright: ignore [reportUnknownVariableType]
 

amt/migrations/env.py

@@ -6,12 +6,14 @@
 from sqlalchemy import engine_from_config, pool
 from sqlalchemy.schema import MetaData
 
+from amt.models.base import Base
+
 config = context.config
 
 if config.config_file_name is not None:
     fileConfig(config.config_file_name)
 
-target_metadata = MetaData()
+target_metadata = Base.metadata
 
 
 def get_url() -> str:

amt/migrations/versions/2c84c4ad5b68_add_reference_to_project_in_task.py

@@ -19,18 +19,14 @@
 
 
 def upgrade() -> None:
-    # ### commands auto generated by Alembic - please adjust! ###
     with op.batch_alter_table("task", schema=None) as batch_op:
         batch_op.add_column(sa.Column("project_id", sa.Integer(), nullable=True))
-        batch_op.create_foreign_key("fk_project_id", "project", ["project_id"], ["id"])
-
-    # ### end Alembic commands ###
+        # Note that the value None below is due to the automatic naming generation from Base.metadata
+        batch_op.create_foreign_key(None, "project", ["project_id"], ["id"])
 
 
 def downgrade() -> None:
-    # ### commands auto generated by Alembic - please adjust! ###
     with op.batch_alter_table("task", schema=None) as batch_op:
-        batch_op.drop_constraint("fk_project_id", type_="foreignkey")
+        # Drop constraint in batch mode lacks the fields to use automatic naming convention
+        batch_op.drop_constraint("fk_task_project_id_project", type_="foreignkey")
         batch_op.drop_column("project_id")
-
-    # ### end Alembic commands ###

amt/migrations/versions/9ce2341f2922_remove_the_status_table.py

@@ -19,28 +19,15 @@
 
 
 def upgrade() -> None:
-    # ### commands auto generated by Alembic - please adjust! ###
-    # fix unnamed foreign keys:
-    #  https://alembic.sqlalchemy.org/en/latest/batch.html#dropping-unnamed-or-named-foreign-key-constraints
-    naming_convention = {
-        "fk":
-            "%(table_name)s_%(column_0_name)s_fkey"
-    }
-    with op.batch_alter_table("task", naming_convention=naming_convention) as batch_op:
+    with op.batch_alter_table("task") as batch_op:
         batch_op.drop_constraint(
-            "task_status_id_fkey", type_="foreignkey")
+            "fk_task_status_id_status", type_="foreignkey")
 
     op.drop_table("status")
     # ### end Alembic commands ###
 
 
 def downgrade() -> None:
-
-    naming_convention = {
-        "fk":
-            "%(table_name)s_%(column_0_name)s_fkey"
-    }
-
     op.create_table(
         "status",
         sa.Column("id", sa.Integer(), nullable=False),
@@ -51,8 +38,5 @@ def downgrade() -> None:
 
     op.execute("UPDATE task SET status_id = NULL")
 
-    # ### commands auto generated by Alembic - please adjust! ###
-    with op.batch_alter_table("task", naming_convention=naming_convention) as batch_op:
-        batch_op.create_foreign_key("task_status_id_fkey", "status", ["status_id"], ["id"])
-
-    # ### end Alembic commands ###
+    with op.batch_alter_table("task") as batch_op:
+        batch_op.create_foreign_key(None, "status", ["status_id"], ["id"])

amt/models/base.py

@@ -1,5 +1,14 @@
+from sqlalchemy import MetaData
 from sqlalchemy.orm import DeclarativeBase
 
 
 class Base(DeclarativeBase):
-    pass
+    metadata = MetaData(
+        naming_convention={
+            "ix": "ix_%(column_0_label)s",
+            "uq": "uq_%(table_name)s_%(column_0_name)s",
+            "ck": "ck_%(table_name)s_`%(constraint_name)s`",
+            "fk": "fk_%(table_name)s_%(column_0_name)s_%(referred_table_name)s",
+            "pk": "pk_%(table_name)s",
+        }
+    )

tests/site/static/templates/test_task_template.py

@@ -18,6 +18,6 @@ def test_template_rendering():
     """
     # todo (robbert) there is probably a better way to clean results than replacing newlines and spaces
     #  and we probably would want a function for it
-    cleaned_results = result.body.decode().replace("\n", "").replace(" ", "")
-    cleaned_expected_result = expected_result.replace("\n", "").replace(" ", "")
+    cleaned_results: str = result.body.decode().replace("\n", "").replace(" ", "")  # pyright: ignore [reportUnknownMemberType, reportUnknownVariableType, reportAttributeAccessIssue]
+    cleaned_expected_result: str = expected_result.replace("\n", "").replace(" ", "")
     assert cleaned_results == cleaned_expected_result