fix: make json_url nullable

Author: tillywoodfield

migrations/versions/3499656b84e7_allow_nullable_json_url.py

@@ -0,0 +1,31 @@
+"""allow nullable json_url
+
+Revision ID: 3499656b84e7
+Revises: 084c39bf418e
+Create Date: 2025-02-11 16:44:30.550413
+
+"""
+
+from typing import Sequence, Union
+
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision: str = "3499656b84e7"
+down_revision: Union[str, None] = "084c39bf418e"
+branch_labels: Union[str, Sequence[str], None] = None
+depends_on: Union[str, Sequence[str], None] = None
+
+
+def upgrade() -> None:
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.alter_column("dataset", "json_url", existing_type=sa.VARCHAR(), nullable=True)
+    # ### end Alembic commands ###
+
+
+def downgrade() -> None:
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.alter_column("dataset", "json_url", existing_type=sa.VARCHAR(), nullable=False)
+    # ### end Alembic commands ###

oc4ids_datastore_pipeline/database.py

@@ -30,7 +30,7 @@ class Dataset(Base):
     publisher_name: Mapped[str] = mapped_column(String)
     license_url: Mapped[Optional[str]] = mapped_column(String, nullable=True)
     license_name: Mapped[Optional[str]] = mapped_column(String, nullable=True)
-    json_url: Mapped[str] = mapped_column(String)
+    json_url: Mapped[str] = mapped_column(String, nullable=True)
     csv_url: Mapped[Optional[str]] = mapped_column(String, nullable=True)
     xlsx_url: Mapped[Optional[str]] = mapped_column(String, nullable=True)
     updated_at: Mapped[datetime.datetime] = mapped_column(DateTime(timezone=True))