sync update from LAN repo (source dhk-1 @ 3b1ec5b4)

This commit is contained in:
gozeros committed 2026-11-05 10:56:34 +08:00
1 parent da6521b6cb
commit 264e2200d6
995 files changed
+148105 -13399

No files matched your search

@@ -15,7 +15,7 @@ from __future__ import annotations
from datetime import UTC, datetime
from sqlalchemy import String
from sqlalchemy import Index, String
from sqlalchemy.orm import Mapped, mapped_column
from deerflow.persistence.base import Base
@@ -24,6 +24,14 @@ from deerflow.persistence.types import BeijingDateTime, PortableLongText
class ParallelAgentTaskRow(Base):
__tablename__ = "parallel_agent_tasks"
__table_args__ = (
Index(
"ix_parallel_agent_tasks_user_updated_id",
"user_id",
"updated_at",
"id",
),
)
id: Mapped[str] = mapped_column(String(64), primary_key=True)
user_id: Mapped[str] = mapped_column(String(64), nullable=False, index=True)
@@ -11,6 +11,7 @@ that belongs to another user is treated as not found.
from __future__ import annotations
import json
from collections.abc import Mapping
from datetime import UTC, datetime
from typing import Any
@@ -44,17 +45,32 @@ class ParallelAgentTaskRepository:
self._sf = session_factory
@staticmethod
def _to_dict(row: ParallelAgentTaskRow, *, with_messages: bool = True) -> dict[str, Any]:
def _to_dict(
row: ParallelAgentTaskRow | Mapping[str, Any],
*,
with_messages: bool = True,
) -> dict[str, Any]:
def value(name: str) -> Any:
if isinstance(row, Mapping):
return row[name]
return getattr(row, name)
created_at = value("created_at")
updated_at = value("updated_at")
data: dict[str, Any] = {
"id": row.id,
"title": row.title,
"brief": row.brief,
"agents": _loads(row.agents) or [],
"created_at": row.created_at.isoformat() if isinstance(row.created_at, datetime) else row.created_at,
"updated_at": row.updated_at.isoformat() if isinstance(row.updated_at, datetime) else row.updated_at,
"id": value("id"),
"title": value("title"),
"brief": value("brief"),
"agents": _loads(value("agents")) or [],
"created_at": (
created_at.isoformat() if isinstance(created_at, datetime) else created_at
),
"updated_at": (
updated_at.isoformat() if isinstance(updated_at, datetime) else updated_at
),
}
if with_messages:
data["messages"] = _loads(row.messages) or []
data["messages"] = _loads(value("messages")) or []
return data
async def list_tasks(
@@ -74,9 +90,22 @@ class ParallelAgentTaskRepository:
and ``offset`` allow those projections to page only their own task
namespace without loading unrelated, potentially large transcripts.
"""
stmt = select(ParallelAgentTaskRow).where(ParallelAgentTaskRow.user_id == user_id)
# History/list callers only need the summary columns. In particular,
# never select the LONGTEXT transcript column unless explicitly asked:
# MySQL otherwise reads and transfers it even though _to_dict omits it.
columns = [
ParallelAgentTaskRow.id,
ParallelAgentTaskRow.title,
ParallelAgentTaskRow.brief,
ParallelAgentTaskRow.agents,
ParallelAgentTaskRow.created_at,
ParallelAgentTaskRow.updated_at,
]
if include_messages:
columns.append(ParallelAgentTaskRow.messages)
stmt = select(*columns).where(ParallelAgentTaskRow.user_id == user_id)
if id_prefix:
stmt = stmt.where(ParallelAgentTaskRow.id.startswith(id_prefix))
stmt = stmt.where(ParallelAgentTaskRow.id.startswith(id_prefix, autoescape=True))
stmt = (
stmt.order_by(
ParallelAgentTaskRow.updated_at.desc(),
@@ -88,8 +117,8 @@ class ParallelAgentTaskRepository:
async with self._sf() as session:
result = await session.execute(stmt)
return [
self._to_dict(row, with_messages=include_messages)
for row in result.scalars()
self._to_dict(row._mapping, with_messages=include_messages)
for row in result
]
async def get_task(self, task_id: str, user_id: str) -> dict[str, Any] | None: