feat: persist DiskSage file lineage ontology

.github/workflows/app-ci.yml

@@ -129,9 +129,14 @@ jobs:
           cd frontend && pnpm run build
 
       - name: Install Playwright Chromium
-        run: cd frontend && pnpm exec playwright install --with-deps chromium
+        timeout-minutes: 10
+        # The hosted Ubuntu runner already provides the system libraries needed by
+        # Chromium. Installing them again with --with-deps can block on apt for
+        # hours and turn a real frontend result into a cancelled check.
+        run: cd frontend && pnpm exec playwright install chromium
 
       - name: Run full product smoke
+        timeout-minutes: 15
         env:
           NEXT_TELEMETRY_DISABLED: "1"
           NARUON_FULL_PRODUCT_BASE_URL: "http://127.0.0.1:3001"

backend/alembic/versions/0018_disksage_file_lineage.py

@@ -0,0 +1,86 @@
+"""persist scoped, encrypted DiskSage file lineage envelopes
+
+Revision ID: 0018_disksage_file_lineage
+Revises: 0017_merge_newsdom_carddav_heads
+Create Date: 2026-08-13 00:00:00.000000
+
+The Rust DiskSage verifier remains authoritative for copy/provider proof. Naruon
+stores the validated envelope for workspace-scoped provenance and exposes only
+the redacted graph projection in list responses.
+"""
+
+from alembic import op
+import sqlalchemy as sa
+
+
+revision = "0018_disksage_file_lineage"
+down_revision = "0017_merge_newsdom_carddav_heads"
+branch_labels = None
+depends_on = None
+
+_TABLE = "disksage_file_lineage_records"
+
+
+def upgrade() -> None:
+    connection = op.get_bind()
+    inspector = sa.inspect(connection)
+    if not inspector.has_table(_TABLE):
+        op.create_table(
+            _TABLE,
+            sa.Column("lineage_record_uid", sa.String(length=96), nullable=False),
+            sa.Column("user_id", sa.String(), nullable=False),
+            sa.Column("organization_id", sa.String(), nullable=True),
+            sa.Column("workspace_id", sa.String(), nullable=False),
+            sa.Column("lineage_fingerprint", sa.String(length=64), nullable=False),
+            sa.Column("envelope_sha256", sa.String(length=64), nullable=False),
+            sa.Column("schema_version", sa.Integer(), nullable=False),
+            sa.Column("schema_kind", sa.String(length=96), nullable=False),
+            sa.Column("source_kind", sa.String(length=64), nullable=False),
+            sa.Column("archive_kind", sa.String(length=64), nullable=False),
+            sa.Column("raw_content_sha256", sa.String(length=64), nullable=False),
+            sa.Column("raw_content_blake3", sa.String(length=64), nullable=False),
+            sa.Column("content_bytes", sa.BigInteger(), nullable=False),
+            sa.Column("ontology_class", sa.String(length=256), nullable=False),
+            sa.Column("ontology_relation_count", sa.Integer(), nullable=False),
+            sa.Column("ontology_predicates", sa.JSON(), nullable=False),
+            sa.Column("provider_name", sa.String(length=32), nullable=False),
+            sa.Column("provider_sync_confirmed", sa.Boolean(), nullable=False),
+            sa.Column(
+                "provider_sync_state",
+                sa.String(length=32),
+                nullable=False,
+                server_default="unknown",
+            ),
+            sa.Column("envelope_json_encrypted", sa.String(), nullable=False),
+            sa.Column("created_at", sa.DateTime(timezone=True), nullable=False),
+            sa.PrimaryKeyConstraint("lineage_record_uid"),
+            sa.UniqueConstraint(
+                "user_id",
+                "workspace_id",
+                "lineage_fingerprint",
+                name="uq_disksage_lineage_workspace_fingerprint",
+            ),
+        )
+
+    op.create_index(
+        "ix_disksage_lineage_scope_time",
+        _TABLE,
+        ["user_id", "organization_id", "workspace_id", "created_at"],
+        if_not_exists=True,
+    )
+    op.create_index(
+        "ix_disksage_lineage_ontology_class",
+        _TABLE,
+        ["workspace_id", "ontology_class"],
+        if_not_exists=True,
+    )
+
+
+def downgrade() -> None:
+    op.drop_index(
+        "ix_disksage_lineage_ontology_class", table_name=_TABLE, if_exists=True
+    )
+    op.drop_index("ix_disksage_lineage_scope_time", table_name=_TABLE, if_exists=True)
+    connection = op.get_bind()
+    if sa.inspect(connection).has_table(_TABLE):
+        op.drop_table(_TABLE)

backend/alembic/versions/0019_disksage_provider_sync_state.py

@@ -0,0 +1,43 @@
+"""persist provider-native sync state for DiskSage lineage envelopes
+
+Revision ID: 0019_disksage_sync_state
+Revises: 0018_disksage_file_lineage
+Create Date: 2026-08-13 00:00:00.000000
+"""
+
+from alembic import op
+import sqlalchemy as sa
+
+
+revision = "0019_disksage_sync_state"
+down_revision = "0018_disksage_file_lineage"
+branch_labels = None
+depends_on = None
+
+_TABLE = "disksage_file_lineage_records"
+
+
+def upgrade() -> None:
+    inspector = sa.inspect(op.get_bind())
+    if inspector.has_table(_TABLE) and not any(
+        column["name"] == "provider_sync_state"
+        for column in inspector.get_columns(_TABLE)
+    ):
+        op.add_column(
+            _TABLE,
+            sa.Column(
+                "provider_sync_state",
+                sa.String(length=32),
+                nullable=False,
+                server_default="unknown",
+            ),
+        )
+
+
+def downgrade() -> None:
+    inspector = sa.inspect(op.get_bind())
+    if inspector.has_table(_TABLE) and any(
+        column["name"] == "provider_sync_state"
+        for column in inspector.get_columns(_TABLE)
+    ):
+        op.drop_column(_TABLE, "provider_sync_state")

backend/alembic/versions/0020_disksage_lineage_column_names.py

@@ -0,0 +1,39 @@
+"""rename ambiguous DiskSage lineage column names
+
+Revision ID: 0020_disksage_columns
+Revises: 0019_disksage_sync_state
+Create Date: 2026-08-13 00:00:00.000000
+"""
+
+from alembic import op
+import sqlalchemy as sa
+
+
+revision = "0020_disksage_columns"
+down_revision = "0019_disksage_sync_state"
+branch_labels = None
+depends_on = None
+
+_TABLE = "disksage_file_lineage_records"
+
+
+def _rename_if_present(inspector: sa.Inspector, old: str, new: str) -> None:
+    columns = {column["name"] for column in inspector.get_columns(_TABLE)}
+    if old in columns and new not in columns:
+        op.alter_column(_TABLE, old, new_column_name=new)
+
+
+def upgrade() -> None:
+    inspector = sa.inspect(op.get_bind())
+    if inspector.has_table(_TABLE):
+        _rename_if_present(inspector, "bytes", "content_bytes")
+        inspector = sa.inspect(op.get_bind())
+        _rename_if_present(inspector, "provider", "provider_name")
+
+
+def downgrade() -> None:
+    # This revision repairs legacy installations that still have the old
+    # single-word names.  Revision 0019 and the canonical 0018 schema both
+    # use content_bytes/provider_name, so a downgrade must not reintroduce
+    # the legacy shape or leave the model and migration chain inconsistent.
+    pass

backend/alembic/versions/0021_disksage_lineage_scope_index.py

@@ -0,0 +1,39 @@
+"""align the DiskSage lineage scope index with the list query"""
+
+from alembic import op
+import sqlalchemy as sa
+
+
+revision = "0021_disksage_scope_idx"
+down_revision = "0020_disksage_columns"
+branch_labels = None
+depends_on = None
+
+_TABLE = "disksage_file_lineage_records"
+_INDEX = "ix_disksage_lineage_scope_time"
+
+
+def upgrade() -> None:
+    inspector = sa.inspect(op.get_bind())
+    if not inspector.has_table(_TABLE):
+        return
+    op.drop_index(_INDEX, table_name=_TABLE, if_exists=True)
+    op.create_index(
+        _INDEX,
+        _TABLE,
+        ["user_id", "workspace_id", "created_at"],
+        if_not_exists=True,
+    )
+
+
+def downgrade() -> None:
+    inspector = sa.inspect(op.get_bind())
+    if not inspector.has_table(_TABLE):
+        return
+    op.drop_index(_INDEX, table_name=_TABLE, if_exists=True)
+    op.create_index(
+        _INDEX,
+        _TABLE,
+        ["user_id", "organization_id", "workspace_id", "created_at"],
+        if_not_exists=True,
+    )

backend/alembic/versions/0022_disksage_organization_lineage.py

@@ -0,0 +1,51 @@
+"""persist path-free DiskSage organization lineage batches"""
+
+from alembic import op
+import sqlalchemy as sa
+
+
+revision = "0022_disksage_org_lineage"
+down_revision = "0021_disksage_scope_idx"
+branch_labels = None
+depends_on = None
+
+_TABLE = "disksage_organization_lineage_records"
+
+
+def upgrade() -> None:
+    inspector = sa.inspect(op.get_bind())
+    if inspector.has_table(_TABLE):
+        return
+    op.create_table(
+        _TABLE,
+        sa.Column("organization_lineage_record_uid", sa.String(length=96), nullable=False),
+        sa.Column("user_id", sa.String(), nullable=False),
+        sa.Column("organization_id", sa.String(), nullable=True),
+        sa.Column("workspace_id", sa.String(), nullable=False),
+        sa.Column("batch_fingerprint_sha256", sa.String(length=64), nullable=False),
+        sa.Column("envelope_sha256", sa.String(length=64), nullable=False),
+        sa.Column("schema_version", sa.Integer(), nullable=False),
+        sa.Column("item_count", sa.Integer(), nullable=False),
+        sa.Column("ontology_classes", sa.JSON(), nullable=False),
+        sa.Column("envelope_json_encrypted", sa.String(), nullable=False),
+        sa.Column("created_at", sa.DateTime(timezone=True), nullable=False),
+        sa.PrimaryKeyConstraint("organization_lineage_record_uid"),
+        sa.UniqueConstraint(
+            "user_id",
+            "workspace_id",
+            "batch_fingerprint_sha256",
+            name="uq_disksage_org_lineage_workspace_fingerprint",
+        ),
+    )
+    op.create_index(
+        "ix_disksage_org_lineage_scope_time",
+        _TABLE,
+        ["user_id", "workspace_id", "created_at"],
+        if_not_exists=True,
+    )
+
+
+def downgrade() -> None:
+    op.drop_index("ix_disksage_org_lineage_scope_time", table_name=_TABLE, if_exists=True)
+    if sa.inspect(op.get_bind()).has_table(_TABLE):
+        op.drop_table(_TABLE)

backend/api/dav.py

@@ -159,6 +159,7 @@ async def _handle_project_propfind(
 @router.api_route(
     "/{path:path}",
     methods=["PROPFIND", "REPORT", "MKCOL", "GET", "PUT", "DELETE", "OPTIONS"],
+    include_in_schema=False,
 )
 async def dav_handler(
     request: Request,