feat(01-03): scaffold Alembic async config and author 0001_initial_schema migration

- backend/alembic.ini: script_location=migrations, sqlalchemy.url=%(DATABASE_MIGRATE_URL)s - backend/migrations/env.py: async_engine_from_config + Base.metadata wiring; runtime os.environ.get("DATABASE_MIGRATE_URL") injection (alembic.ini interpolation does not read OS env directly) - backend/migrations/versions/0001_initial_schema.py: creates all 11 tables in dependency order with correct FKs, indexes, and named constraints - documents.user_id is nullable=True per D-03; Phase 2 adds NOT NULL - Ends with GRANT + ALTER DEFAULT PRIVILEGES for docuvault_app (Pitfall 4) - Also grants USAGE/SELECT on sequences (audit_log.id autoincrement) - downgrade() drops all tables in reverse dependency order
2026-05-22 09:20:49 +02:00
parent 3e1fcd69b5
commit 75ea7ef106
5 changed files with 536 additions and 0 deletions
@@ -0,0 +1,257 @@
+"""Initial v1 schema — all 11 DocuVault tables.
+
+Revision ID: 0001
+Revises:
+Create Date: 2026-05-22
+
+Creates tables in dependency order:
+  1. users              (no FK dependencies)
+  2. groups             (no FK dependencies — D-02 stub)
+  3. quotas             (FK -> users)
+  4. refresh_tokens     (FK -> users)
+  5. folders            (FK -> users, self-referential FK -> folders)
+  6. topics             (FK -> users)
+  7. documents          (FK -> users [nullable, D-03], FK -> folders)
+  8. document_topics    (FK -> documents, topics)
+  9. shares             (FK -> documents, users x2)
+ 10. audit_log          (FK -> users x2)
+ 11. cloud_connections  (FK -> users)
+
+Pitfall 4 note: ALTER DEFAULT PRIVILEGES is required so future migrations
+(applied by docuvault_migrate) automatically grant access to docuvault_app.
+The docuvault_app user is created in docker/postgres/initdb.d/01-init-users.sql
+with CONNECT but no table privileges — this migration grants them.
+"""
+from __future__ import annotations
+
+import sqlalchemy as sa
+from sqlalchemy.dialects import postgresql
+from alembic import op
+
+# revision identifiers, used by Alembic.
+revision = "0001"
+down_revision = None
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    # ── 1. users ──────────────────────────────────────────────────────────────
+    op.create_table("users",
+        sa.Column("id", postgresql.UUID(as_uuid=True), nullable=False),
+        sa.Column("handle", sa.String(), nullable=False),
+        sa.Column("email", sa.String(), nullable=False),
+        sa.Column("password_hash", sa.Text(), nullable=False),
+        sa.Column("totp_secret", sa.Text(), nullable=True),
+        sa.Column("totp_enabled", sa.Boolean(), nullable=False, server_default="false"),
+        sa.Column("role", sa.String(), nullable=False, server_default="user"),
+        sa.Column("is_active", sa.Boolean(), nullable=False, server_default="true"),
+        sa.Column("ai_provider", sa.Text(), nullable=True),
+        sa.Column("ai_model", sa.Text(), nullable=True),
+        sa.Column("default_storage_backend", sa.String(), nullable=False, server_default="minio"),
+        sa.Column("created_at", sa.TIMESTAMP(timezone=True), nullable=False, server_default=sa.text("now()")),
+        sa.PrimaryKeyConstraint("id"),
+        sa.UniqueConstraint("handle"),
+        sa.UniqueConstraint("email"),
+    )
+
+    # ── 2. groups (D-02 stub — v2 feature) ───────────────────────────────────
+    op.create_table("groups",
+        sa.Column("id", postgresql.UUID(as_uuid=True), nullable=False),
+        sa.Column("name", sa.Text(), nullable=False),
+        sa.Column("created_at", sa.TIMESTAMP(timezone=True), nullable=False, server_default=sa.text("now()")),
+        sa.PrimaryKeyConstraint("id"),
+        sa.UniqueConstraint("name"),
+    )
+
+    # ── 3. quotas ─────────────────────────────────────────────────────────────
+    op.create_table("quotas",
+        sa.Column("user_id", postgresql.UUID(as_uuid=True), nullable=False),
+        # 100 MB default free-tier quota
+        sa.Column("limit_bytes", sa.BigInteger(), nullable=False, server_default="104857600"),
+        sa.Column("used_bytes", sa.BigInteger(), nullable=False, server_default="0"),
+        sa.ForeignKeyConstraint(["user_id"], ["users.id"], ondelete="CASCADE"),
+        sa.PrimaryKeyConstraint("user_id"),
+    )
+
+    # ── 4. refresh_tokens ─────────────────────────────────────────────────────
+    op.create_table("refresh_tokens",
+        sa.Column("id", postgresql.UUID(as_uuid=True), nullable=False),
+        sa.Column("user_id", postgresql.UUID(as_uuid=True), nullable=False),
+        sa.Column("token_hash", sa.Text(), nullable=False),
+        sa.Column("expires_at", sa.TIMESTAMP(timezone=True), nullable=False),
+        sa.Column("revoked", sa.Boolean(), nullable=False, server_default="false"),
+        sa.Column("created_at", sa.TIMESTAMP(timezone=True), nullable=False, server_default=sa.text("now()")),
+        sa.ForeignKeyConstraint(["user_id"], ["users.id"], ondelete="CASCADE"),
+        sa.PrimaryKeyConstraint("id"),
+        sa.UniqueConstraint("token_hash"),
+    )
+    op.create_index("ix_refresh_tokens_user_revoked", "refresh_tokens", ["user_id", "revoked"])
+
+    # ── 5. folders ────────────────────────────────────────────────────────────
+    op.create_table("folders",
+        sa.Column("id", postgresql.UUID(as_uuid=True), nullable=False),
+        sa.Column("user_id", postgresql.UUID(as_uuid=True), nullable=False),
+        sa.Column("parent_id", postgresql.UUID(as_uuid=True), nullable=True),
+        sa.Column("name", sa.Text(), nullable=False),
+        sa.Column("created_at", sa.TIMESTAMP(timezone=True), nullable=False, server_default=sa.text("now()")),
+        sa.ForeignKeyConstraint(["user_id"], ["users.id"], ondelete="CASCADE"),
+        sa.ForeignKeyConstraint(["parent_id"], ["folders.id"], ondelete="CASCADE"),
+        sa.PrimaryKeyConstraint("id"),
+        sa.UniqueConstraint("user_id", "parent_id", "name", name="uq_folders_user_parent_name"),
+    )
+
+    # ── 6. topics ─────────────────────────────────────────────────────────────
+    op.create_table("topics",
+        sa.Column("id", postgresql.UUID(as_uuid=True), nullable=False),
+        sa.Column("user_id", postgresql.UUID(as_uuid=True), nullable=True),
+        sa.Column("name", sa.Text(), nullable=False),
+        sa.Column("description", sa.Text(), nullable=False, server_default=""),
+        sa.Column("color", sa.String(7), nullable=False, server_default="#6366f1"),
+        sa.ForeignKeyConstraint(["user_id"], ["users.id"], ondelete="CASCADE"),
+        sa.PrimaryKeyConstraint("id"),
+        sa.UniqueConstraint("user_id", "name", name="uq_topics_user_name"),
+    )
+
+    # ── 7. documents ──────────────────────────────────────────────────────────
+    op.create_table("documents",
+        sa.Column("id", postgresql.UUID(as_uuid=True), nullable=False),
+        # D-03: user_id is nullable in Phase 1 — no auth yet.
+        # Phase 2 migration adds NOT NULL constraint.
+        sa.Column("user_id", postgresql.UUID(as_uuid=True), nullable=True),
+        sa.Column("folder_id", postgresql.UUID(as_uuid=True), nullable=True),
+        # original human-readable filename — stored in DB only, never in the MinIO key
+        sa.Column("filename", sa.Text(), nullable=False),
+        # MinIO object key: {user_id}/{document_id}/{uuid4()}{ext}
+        sa.Column("object_key", sa.Text(), nullable=False),
+        sa.Column("content_type", sa.Text(), nullable=False),
+        sa.Column("size_bytes", sa.BigInteger(), nullable=False, server_default="0"),
+        sa.Column("storage_backend", sa.String(), nullable=False, server_default="minio"),
+        sa.Column("extracted_text", sa.Text(), nullable=True),
+        sa.Column("status", sa.String(), nullable=False, server_default="pending"),
+        sa.Column("created_at", sa.TIMESTAMP(timezone=True), nullable=False, server_default=sa.text("now()")),
+        sa.Column("updated_at", sa.TIMESTAMP(timezone=True), nullable=False, server_default=sa.text("now()")),
+        sa.ForeignKeyConstraint(["user_id"], ["users.id"], ondelete="CASCADE"),
+        sa.ForeignKeyConstraint(["folder_id"], ["folders.id"], ondelete="SET NULL"),
+        sa.PrimaryKeyConstraint("id"),
+    )
+    op.create_index("ix_documents_user_folder", "documents", ["user_id", "folder_id"])
+    op.create_index("ix_documents_user_created", "documents", ["user_id", "created_at"])
+
+    # ── 8. document_topics ────────────────────────────────────────────────────
+    op.create_table("document_topics",
+        sa.Column("document_id", postgresql.UUID(as_uuid=True), nullable=False),
+        sa.Column("topic_id", postgresql.UUID(as_uuid=True), nullable=False),
+        sa.ForeignKeyConstraint(["document_id"], ["documents.id"], ondelete="CASCADE"),
+        sa.ForeignKeyConstraint(["topic_id"], ["topics.id"], ondelete="CASCADE"),
+        sa.PrimaryKeyConstraint("document_id", "topic_id"),
+    )
+
+    # ── 9. shares ─────────────────────────────────────────────────────────────
+    op.create_table("shares",
+        sa.Column("id", postgresql.UUID(as_uuid=True), nullable=False),
+        sa.Column("document_id", postgresql.UUID(as_uuid=True), nullable=False),
+        sa.Column("owner_id", postgresql.UUID(as_uuid=True), nullable=False),
+        sa.Column("recipient_id", postgresql.UUID(as_uuid=True), nullable=False),
+        sa.Column("permission", sa.String(), nullable=False, server_default="view"),
+        sa.Column("created_at", sa.TIMESTAMP(timezone=True), nullable=False, server_default=sa.text("now()")),
+        sa.ForeignKeyConstraint(["document_id"], ["documents.id"], ondelete="CASCADE"),
+        sa.ForeignKeyConstraint(["owner_id"], ["users.id"], ondelete="CASCADE"),
+        sa.ForeignKeyConstraint(["recipient_id"], ["users.id"], ondelete="CASCADE"),
+        sa.PrimaryKeyConstraint("id"),
+        sa.UniqueConstraint("document_id", "recipient_id", name="uq_shares_document_recipient"),
+    )
+    op.create_index("ix_shares_recipient", "shares", ["recipient_id"])
+
+    # ── 10. audit_log ─────────────────────────────────────────────────────────
+    op.create_table("audit_log",
+        sa.Column("id", sa.Integer(), autoincrement=True, nullable=False),
+        sa.Column("user_id", postgresql.UUID(as_uuid=True), nullable=True),
+        sa.Column("actor_id", postgresql.UUID(as_uuid=True), nullable=True),
+        sa.Column("event_type", sa.Text(), nullable=False),
+        sa.Column("resource_id", postgresql.UUID(as_uuid=True), nullable=True),
+        sa.Column("ip_address", sa.dialects.postgresql.INET(), nullable=True),
+        # DB column name is "metadata"; ORM uses "metadata_" to avoid reserved-attr conflict
+        sa.Column("metadata", sa.dialects.postgresql.JSONB(), nullable=True),
+        sa.Column("created_at", sa.TIMESTAMP(timezone=True), nullable=False, server_default=sa.text("now()")),
+        sa.ForeignKeyConstraint(["user_id"], ["users.id"], ondelete="SET NULL"),
+        sa.ForeignKeyConstraint(["actor_id"], ["users.id"], ondelete="SET NULL"),
+        sa.PrimaryKeyConstraint("id"),
+    )
+    op.create_index("ix_audit_user_created", "audit_log", ["user_id", "created_at"])
+    op.create_index("ix_audit_event_created", "audit_log", ["event_type", "created_at"])
+
+    # ── 11. cloud_connections ─────────────────────────────────────────────────
+    op.create_table("cloud_connections",
+        sa.Column("id", postgresql.UUID(as_uuid=True), nullable=False),
+        sa.Column("user_id", postgresql.UUID(as_uuid=True), nullable=False),
+        sa.Column("provider", sa.String(), nullable=False),
+        sa.Column("display_name", sa.Text(), nullable=False),
+        sa.Column("credentials_enc", sa.Text(), nullable=False),
+        sa.Column("status", sa.String(), nullable=False, server_default="ACTIVE"),
+        sa.Column("connected_at", sa.TIMESTAMP(timezone=True), nullable=False, server_default=sa.text("now()")),
+        sa.ForeignKeyConstraint(["user_id"], ["users.id"], ondelete="CASCADE"),
+        sa.PrimaryKeyConstraint("id"),
+    )
+    op.create_index("ix_cloud_connections_user", "cloud_connections", ["user_id"])
+
+    # ── Privilege grants ───────────────────────────────────────────────────────
+    # Pitfall 4: ALTER DEFAULT PRIVILEGES is required so future migrations
+    # (applied by docuvault_migrate user) automatically grant DML access to
+    # docuvault_app on any tables created afterward.
+    #
+    # The docuvault_app user is created in docker/postgres/initdb.d/01-init-users.sql
+    # with CONNECT privilege only — no table-level access. These two grants establish:
+    #   (a) immediate access to all tables/sequences created by THIS migration
+    #   (b) automatic access to all tables/sequences created by FUTURE migrations
+    #       run by the same docuvault_migrate user
+    op.execute("GRANT SELECT, INSERT, UPDATE, DELETE ON ALL TABLES IN SCHEMA public TO docuvault_app;")
+    op.execute("ALTER DEFAULT PRIVILEGES IN SCHEMA public GRANT SELECT, INSERT, UPDATE, DELETE ON TABLES TO docuvault_app;")
+    # Sequences grant is required because audit_log.id uses a sequence (autoincrement).
+    # docuvault_app must be able to call nextval() on any sequence.
+    op.execute("GRANT USAGE, SELECT ON ALL SEQUENCES IN SCHEMA public TO docuvault_app;")
+    op.execute("ALTER DEFAULT PRIVILEGES IN SCHEMA public GRANT USAGE, SELECT ON SEQUENCES TO docuvault_app;")
+
+
+def downgrade() -> None:
+    # Drop indexes first, then tables in reverse dependency order.
+
+    # cloud_connections
+    op.drop_index("ix_cloud_connections_user", table_name="cloud_connections")
+    op.drop_table("cloud_connections")
+
+    # audit_log
+    op.drop_index("ix_audit_event_created", table_name="audit_log")
+    op.drop_index("ix_audit_user_created", table_name="audit_log")
+    op.drop_table("audit_log")
+
+    # shares
+    op.drop_index("ix_shares_recipient", table_name="shares")
+    op.drop_table("shares")
+
+    # document_topics
+    op.drop_table("document_topics")
+
+    # documents
+    op.drop_index("ix_documents_user_created", table_name="documents")
+    op.drop_index("ix_documents_user_folder", table_name="documents")
+    op.drop_table("documents")
+
+    # topics
+    op.drop_table("topics")
+
+    # folders
+    op.drop_table("folders")
+
+    # refresh_tokens
+    op.drop_index("ix_refresh_tokens_user_revoked", table_name="refresh_tokens")
+    op.drop_table("refresh_tokens")
+
+    # quotas
+    op.drop_table("quotas")
+
+    # groups (D-02 stub)
+    op.drop_table("groups")
+
+    # users (last — all other tables depend on it)
+    op.drop_table("users")