deerflow-code/offline-backend-20260512/backend/packages/harness/deerflow/persistence/llmwiki_index/model.py
2026-09-07 18:24:55 +08:00

95 lines
6.3 KiB
Python

"""Portable ORM rows for the DeerFlow-local WeKnora Wiki mirror/index."""
from __future__ import annotations
from datetime import UTC, datetime
from sqlalchemy import BigInteger, Boolean, ForeignKey, Index, Integer, LargeBinary, String, Text, UniqueConstraint, false
from sqlalchemy.orm import Mapped, mapped_column
from deerflow.persistence.base import Base
from deerflow.persistence.types import BeijingDateTime, PortableJSON, PortableLongText
def _now() -> datetime:
return datetime.now(UTC)
class LlmWikiWikiPageRow(Base):
__tablename__ = "llmwiki_wiki_pages"
__table_args__ = (
UniqueConstraint("knowledge_base_mapping_id", "slug_hash", name="uq_llmwiki_wiki_page_mapping_slug_hash"),
Index("ix_llmwiki_wiki_page_mapping_visibility", "knowledge_base_mapping_id", "status", "index_status", "is_remote_deleted"),
)
id: Mapped[str] = mapped_column(String(36), primary_key=True)
knowledge_base_mapping_id: Mapped[str] = mapped_column(String(36), ForeignKey("llmwiki_knowledge_bases.id", ondelete="CASCADE"), nullable=False, index=True)
remote_page_id: Mapped[str | None] = mapped_column(String(128), nullable=True, index=True)
slug: Mapped[str] = mapped_column(String(1024), nullable=False)
slug_hash: Mapped[str] = mapped_column(String(64), nullable=False)
title: Mapped[str] = mapped_column(String(512), nullable=False, default="", index=True)
summary: Mapped[str] = mapped_column(PortableLongText(), nullable=False, default="")
content_md: Mapped[str] = mapped_column(PortableLongText(), nullable=False, default="")
page_type: Mapped[str] = mapped_column(String(32), nullable=False, default="article", index=True)
status: Mapped[str] = mapped_column(String(32), nullable=False, default="published", index=True)
aliases: Mapped[list] = mapped_column(PortableJSON(), nullable=False, default=list)
category_path: Mapped[list] = mapped_column(PortableJSON(), nullable=False, default=list)
parent_slug: Mapped[str | None] = mapped_column(String(1024), nullable=True)
wiki_path: Mapped[str | None] = mapped_column(String(1024), nullable=True)
source_refs: Mapped[list] = mapped_column(PortableJSON(), nullable=False, default=list)
chunk_refs: Mapped[list] = mapped_column(PortableJSON(), nullable=False, default=list)
page_version: Mapped[int] = mapped_column(Integer, nullable=False, default=0, index=True)
content_hash: Mapped[str] = mapped_column(String(64), nullable=False, index=True)
remote_created_at: Mapped[datetime | None] = mapped_column(BeijingDateTime(), nullable=True)
remote_updated_at: Mapped[datetime | None] = mapped_column(BeijingDateTime(), nullable=True, index=True)
index_status: Mapped[str] = mapped_column(String(24), nullable=False, default="pending", index=True)
indexed_fingerprint: Mapped[str | None] = mapped_column(String(64), nullable=True)
indexed_at: Mapped[datetime | None] = mapped_column(BeijingDateTime(), nullable=True)
last_error: Mapped[str | None] = mapped_column(Text, nullable=True)
is_remote_deleted: Mapped[bool] = mapped_column(Boolean, nullable=False, default=False, server_default=false(), index=True)
last_seen_sync_id: Mapped[str | None] = mapped_column(String(36), nullable=True, index=True)
created_at: Mapped[datetime] = mapped_column(BeijingDateTime(), nullable=False, default=_now)
updated_at: Mapped[datetime] = mapped_column(BeijingDateTime(), nullable=False, default=_now, onupdate=_now, index=True)
class LlmWikiWikiVectorRow(Base):
__tablename__ = "llmwiki_wiki_vectors"
__table_args__ = (
UniqueConstraint("page_id", "section_index", "embedding_fingerprint", name="uq_llmwiki_wiki_vector_page_section_fingerprint"),
Index("ix_llmwiki_wiki_vector_snapshot", "knowledge_base_mapping_id", "embedding_fingerprint"),
)
id: Mapped[str] = mapped_column(String(36), primary_key=True)
page_id: Mapped[str] = mapped_column(String(36), ForeignKey("llmwiki_wiki_pages.id", ondelete="CASCADE"), nullable=False, index=True)
knowledge_base_mapping_id: Mapped[str] = mapped_column(String(36), nullable=False, index=True)
section_index: Mapped[int] = mapped_column(Integer, nullable=False)
heading: Mapped[str | None] = mapped_column(String(1024), nullable=True)
section_content: Mapped[str] = mapped_column(PortableLongText(), nullable=False)
content_hash: Mapped[str] = mapped_column(String(64), nullable=False, index=True)
embedding_fingerprint: Mapped[str] = mapped_column(String(64), nullable=False, index=True)
embedding_dimensions: Mapped[int] = mapped_column(Integer, nullable=False)
vector_blob: Mapped[bytes] = mapped_column(LargeBinary, nullable=False)
created_at: Mapped[datetime] = mapped_column(BeijingDateTime(), nullable=False, default=_now)
class LlmWikiWikiSyncStateRow(Base):
__tablename__ = "llmwiki_wiki_sync_states"
knowledge_base_mapping_id: Mapped[str] = mapped_column(String(36), ForeignKey("llmwiki_knowledge_bases.id", ondelete="CASCADE"), primary_key=True)
state: Mapped[str] = mapped_column(String(24), nullable=False, default="idle")
active_sync_id: Mapped[str | None] = mapped_column(String(36), nullable=True)
lease_owner: Mapped[str | None] = mapped_column(String(128), nullable=True)
lease_expires_at: Mapped[datetime | None] = mapped_column(BeijingDateTime(), nullable=True)
last_started_at: Mapped[datetime | None] = mapped_column(BeijingDateTime(), nullable=True)
last_completed_at: Mapped[datetime | None] = mapped_column(BeijingDateTime(), nullable=True)
last_success_at: Mapped[datetime | None] = mapped_column(BeijingDateTime(), nullable=True)
last_error: Mapped[str | None] = mapped_column(Text, nullable=True)
remote_page_count: Mapped[int] = mapped_column(Integer, nullable=False, default=0)
local_page_count: Mapped[int] = mapped_column(Integer, nullable=False, default=0)
ready_page_count: Mapped[int] = mapped_column(Integer, nullable=False, default=0)
vector_count: Mapped[int] = mapped_column(Integer, nullable=False, default=0)
failed_page_count: Mapped[int] = mapped_column(Integer, nullable=False, default=0)
index_revision: Mapped[int] = mapped_column(BigInteger, nullable=False, default=0)
embedding_fingerprint: Mapped[str | None] = mapped_column(String(64), nullable=True)
updated_at: Mapped[datetime] = mapped_column(BeijingDateTime(), nullable=False, default=_now, onupdate=_now)