PMFlow/backend/alembic/versions/20260302_add_project_knowle...

216 lines
12 KiB
Python

"""add project knowledge graph tables
Revision ID: add_project_knowledge_graph_tables
Revises: add_qa_sessions_and_turns
Create Date: 2026-03-02
"""
from __future__ import annotations
from alembic import op
import sqlalchemy as sa
try:
from sqlalchemy.dialects.postgresql import JSONB
except Exception: # pragma: no cover
JSONB = sa.JSON
# revision identifiers, used by Alembic.
revision = "add_project_knowledge_graph_tables"
down_revision = "add_qa_sessions_and_turns"
branch_labels = None
depends_on = None
def _json_type(bind):
return JSONB() if bind.dialect.name == "postgresql" else sa.JSON()
def _table_exists(bind, table_name: str) -> bool:
inspector = sa.inspect(bind)
return inspector.has_table(table_name)
def _index_exists(bind, table_name: str, index_name: str) -> bool:
inspector = sa.inspect(bind)
indexes = inspector.get_indexes(table_name)
return any(idx.get("name") == index_name for idx in indexes)
def _create_index_if_not_exists(bind, index_name: str, table_name: str, columns: list[str]) -> None:
if not _index_exists(bind, table_name, index_name):
op.create_index(index_name, table_name, columns)
def _drop_index_if_exists(bind, index_name: str, table_name: str) -> None:
if _index_exists(bind, table_name, index_name):
op.drop_index(index_name, table_name=table_name)
def upgrade() -> None:
bind = op.get_bind()
if not _table_exists(bind, "kg_entities"):
op.create_table(
"kg_entities",
sa.Column("id", sa.Integer(), primary_key=True, autoincrement=True),
sa.Column("tenant_id", sa.BigInteger(), nullable=False, server_default="1"),
sa.Column("created_at", sa.TIMESTAMP(), nullable=False, server_default=sa.func.now()),
sa.Column("updated_at", sa.TIMESTAMP(), nullable=False, server_default=sa.func.now(), onupdate=sa.func.now()),
sa.Column("entity_type", sa.String(length=40), nullable=False),
sa.Column("entity_key", sa.String(length=255), nullable=False),
sa.Column("display_name", sa.String(length=255), nullable=False),
sa.Column("payload", _json_type(bind), nullable=True),
sa.Column("source_type", sa.String(length=32), nullable=False, server_default="system"),
sa.Column("source_ref", sa.String(length=255), nullable=True),
sa.Column("last_seen_at", sa.TIMESTAMP(), nullable=False, server_default=sa.func.now()),
sa.UniqueConstraint("tenant_id", "entity_type", "entity_key", name="uq_kg_entities_tenant_type_key"),
)
_create_index_if_not_exists(bind, "ix_kg_entities_tenant_id", "kg_entities", ["tenant_id"])
_create_index_if_not_exists(bind, "ix_kg_entities_entity_type", "kg_entities", ["entity_type"])
_create_index_if_not_exists(bind, "ix_kg_entities_entity_key", "kg_entities", ["entity_key"])
_create_index_if_not_exists(bind, "ix_kg_entities_source_type", "kg_entities", ["source_type"])
_create_index_if_not_exists(bind, "ix_kg_entities_last_seen_at", "kg_entities", ["last_seen_at"])
if not _table_exists(bind, "kg_relations"):
op.create_table(
"kg_relations",
sa.Column("id", sa.Integer(), primary_key=True, autoincrement=True),
sa.Column("tenant_id", sa.BigInteger(), nullable=False, server_default="1"),
sa.Column("created_at", sa.TIMESTAMP(), nullable=False, server_default=sa.func.now()),
sa.Column("updated_at", sa.TIMESTAMP(), nullable=False, server_default=sa.func.now(), onupdate=sa.func.now()),
sa.Column("relation_type", sa.String(length=40), nullable=False),
sa.Column("subject_entity_id", sa.Integer(), sa.ForeignKey("kg_entities.id", ondelete="CASCADE"), nullable=False),
sa.Column("object_entity_id", sa.Integer(), sa.ForeignKey("kg_entities.id", ondelete="CASCADE"), nullable=False),
sa.Column("weight", sa.Float(), nullable=False, server_default="1"),
sa.Column("payload", _json_type(bind), nullable=True),
sa.Column("source_type", sa.String(length=32), nullable=False, server_default="system"),
sa.Column("source_ref", sa.String(length=255), nullable=True),
sa.UniqueConstraint(
"tenant_id",
"relation_type",
"subject_entity_id",
"object_entity_id",
name="uq_kg_relations_tenant_type_subject_object",
),
)
_create_index_if_not_exists(bind, "ix_kg_relations_tenant_id", "kg_relations", ["tenant_id"])
_create_index_if_not_exists(bind, "ix_kg_relations_relation_type", "kg_relations", ["relation_type"])
_create_index_if_not_exists(bind, "ix_kg_relations_subject_entity_id", "kg_relations", ["subject_entity_id"])
_create_index_if_not_exists(bind, "ix_kg_relations_object_entity_id", "kg_relations", ["object_entity_id"])
_create_index_if_not_exists(bind, "ix_kg_relations_source_type", "kg_relations", ["source_type"])
if not _table_exists(bind, "kg_evidences"):
op.create_table(
"kg_evidences",
sa.Column("id", sa.Integer(), primary_key=True, autoincrement=True),
sa.Column("tenant_id", sa.BigInteger(), nullable=False, server_default="1"),
sa.Column("created_at", sa.TIMESTAMP(), nullable=False, server_default=sa.func.now()),
sa.Column("updated_at", sa.TIMESTAMP(), nullable=False, server_default=sa.func.now(), onupdate=sa.func.now()),
sa.Column("entity_id", sa.Integer(), sa.ForeignKey("kg_entities.id", ondelete="CASCADE"), nullable=True),
sa.Column("relation_id", sa.Integer(), sa.ForeignKey("kg_relations.id", ondelete="CASCADE"), nullable=True),
sa.Column("evidence_type", sa.String(length=40), nullable=False),
sa.Column("content_hash", sa.String(length=64), nullable=False),
sa.Column("content_excerpt", sa.Text(), nullable=False),
sa.Column("source_type", sa.String(length=32), nullable=False, server_default="system"),
sa.Column("source_ref", sa.String(length=255), nullable=True),
sa.Column("metadata", _json_type(bind), nullable=True),
sa.Column("occurred_at", sa.TIMESTAMP(), nullable=False, server_default=sa.func.now()),
)
_create_index_if_not_exists(bind, "ix_kg_evidences_tenant_id", "kg_evidences", ["tenant_id"])
_create_index_if_not_exists(bind, "ix_kg_evidences_entity_id", "kg_evidences", ["entity_id"])
_create_index_if_not_exists(bind, "ix_kg_evidences_relation_id", "kg_evidences", ["relation_id"])
_create_index_if_not_exists(bind, "ix_kg_evidences_evidence_type", "kg_evidences", ["evidence_type"])
_create_index_if_not_exists(bind, "ix_kg_evidences_content_hash", "kg_evidences", ["content_hash"])
_create_index_if_not_exists(bind, "ix_kg_evidences_source_type", "kg_evidences", ["source_type"])
_create_index_if_not_exists(bind, "ix_kg_evidences_occurred_at", "kg_evidences", ["occurred_at"])
if not _table_exists(bind, "ingest_events"):
op.create_table(
"ingest_events",
sa.Column("id", sa.Integer(), primary_key=True, autoincrement=True),
sa.Column("tenant_id", sa.BigInteger(), nullable=False, server_default="1"),
sa.Column("created_at", sa.TIMESTAMP(), nullable=False, server_default=sa.func.now()),
sa.Column("updated_at", sa.TIMESTAMP(), nullable=False, server_default=sa.func.now(), onupdate=sa.func.now()),
sa.Column("channel", sa.String(length=32), nullable=False),
sa.Column("event_key", sa.String(length=128), nullable=False),
sa.Column("status", sa.String(length=20), nullable=False, server_default="pending"),
sa.Column("payload", _json_type(bind), nullable=True),
sa.Column("retry_count", sa.Integer(), nullable=False, server_default="0"),
sa.Column("error_message", sa.Text(), nullable=True),
sa.Column("processed_at", sa.TIMESTAMP(), nullable=True),
sa.UniqueConstraint("tenant_id", "channel", "event_key", name="uq_ingest_events_tenant_channel_key"),
)
_create_index_if_not_exists(bind, "ix_ingest_events_tenant_id", "ingest_events", ["tenant_id"])
_create_index_if_not_exists(bind, "ix_ingest_events_channel", "ingest_events", ["channel"])
_create_index_if_not_exists(bind, "ix_ingest_events_event_key", "ingest_events", ["event_key"])
_create_index_if_not_exists(bind, "ix_ingest_events_status", "ingest_events", ["status"])
_create_index_if_not_exists(bind, "ix_ingest_events_processed_at", "ingest_events", ["processed_at"])
if not _table_exists(bind, "ingest_offsets"):
op.create_table(
"ingest_offsets",
sa.Column("id", sa.Integer(), primary_key=True, autoincrement=True),
sa.Column("tenant_id", sa.BigInteger(), nullable=False, server_default="1"),
sa.Column("created_at", sa.TIMESTAMP(), nullable=False, server_default=sa.func.now()),
sa.Column("updated_at", sa.TIMESTAMP(), nullable=False, server_default=sa.func.now(), onupdate=sa.func.now()),
sa.Column("channel", sa.String(length=32), nullable=False),
sa.Column("offset_key", sa.String(length=128), nullable=False),
sa.Column("offset_value", sa.String(length=255), nullable=False),
sa.Column("payload", _json_type(bind), nullable=True),
sa.UniqueConstraint("tenant_id", "channel", "offset_key", name="uq_ingest_offsets_tenant_channel_key"),
)
_create_index_if_not_exists(bind, "ix_ingest_offsets_tenant_id", "ingest_offsets", ["tenant_id"])
_create_index_if_not_exists(bind, "ix_ingest_offsets_channel", "ingest_offsets", ["channel"])
_create_index_if_not_exists(bind, "ix_ingest_offsets_offset_key", "ingest_offsets", ["offset_key"])
def downgrade() -> None:
bind = op.get_bind()
if _table_exists(bind, "ingest_offsets"):
_drop_index_if_exists(bind, "ix_ingest_offsets_offset_key", "ingest_offsets")
_drop_index_if_exists(bind, "ix_ingest_offsets_channel", "ingest_offsets")
_drop_index_if_exists(bind, "ix_ingest_offsets_tenant_id", "ingest_offsets")
op.drop_table("ingest_offsets")
if _table_exists(bind, "ingest_events"):
_drop_index_if_exists(bind, "ix_ingest_events_processed_at", "ingest_events")
_drop_index_if_exists(bind, "ix_ingest_events_status", "ingest_events")
_drop_index_if_exists(bind, "ix_ingest_events_event_key", "ingest_events")
_drop_index_if_exists(bind, "ix_ingest_events_channel", "ingest_events")
_drop_index_if_exists(bind, "ix_ingest_events_tenant_id", "ingest_events")
op.drop_table("ingest_events")
if _table_exists(bind, "kg_evidences"):
_drop_index_if_exists(bind, "ix_kg_evidences_occurred_at", "kg_evidences")
_drop_index_if_exists(bind, "ix_kg_evidences_source_type", "kg_evidences")
_drop_index_if_exists(bind, "ix_kg_evidences_content_hash", "kg_evidences")
_drop_index_if_exists(bind, "ix_kg_evidences_evidence_type", "kg_evidences")
_drop_index_if_exists(bind, "ix_kg_evidences_relation_id", "kg_evidences")
_drop_index_if_exists(bind, "ix_kg_evidences_entity_id", "kg_evidences")
_drop_index_if_exists(bind, "ix_kg_evidences_tenant_id", "kg_evidences")
op.drop_table("kg_evidences")
if _table_exists(bind, "kg_relations"):
_drop_index_if_exists(bind, "ix_kg_relations_source_type", "kg_relations")
_drop_index_if_exists(bind, "ix_kg_relations_object_entity_id", "kg_relations")
_drop_index_if_exists(bind, "ix_kg_relations_subject_entity_id", "kg_relations")
_drop_index_if_exists(bind, "ix_kg_relations_relation_type", "kg_relations")
_drop_index_if_exists(bind, "ix_kg_relations_tenant_id", "kg_relations")
op.drop_table("kg_relations")
if _table_exists(bind, "kg_entities"):
_drop_index_if_exists(bind, "ix_kg_entities_last_seen_at", "kg_entities")
_drop_index_if_exists(bind, "ix_kg_entities_source_type", "kg_entities")
_drop_index_if_exists(bind, "ix_kg_entities_entity_key", "kg_entities")
_drop_index_if_exists(bind, "ix_kg_entities_entity_type", "kg_entities")
_drop_index_if_exists(bind, "ix_kg_entities_tenant_id", "kg_entities")
op.drop_table("kg_entities")