董宏凯andClaude Opus 4.8
3ab674e438
Step2 提速: - 每席位推理深度覆盖(seat_mode): 业务链条逐席位可配 flash/thinking/pro/ultra, 覆盖链条级 seatMode;前台三处 dispatch + 后台 job 透传, run_seat 用 _SEAT_MODE_PARAMS 解析 - recommend 后端批内并行: engine._run_recommend 由串行 for 改 asyncio.gather(对齐前端) - 研讨前并行取数两段式(gather_first, 业务链条 opt-in): 研讨前全席位并行只取数, 产出作为共享资料注入研讨轮(_seat_message gathered + 前台 runGatherPhase); 新增 gather_first 列(migration 20260620_02 + idempotent SQL) - 削减重复广播: suppressPreBroadcast 扩到 recommend(round>=2)/chain(i>0)/chain收口, 保留首轮背景打底,省 N 次读写大 thread 方案总结报告效果: - summary 加"意图对齐+缺口核对+防幻觉护栏": SOUL + 后端 build_summary_prompt + 前端 buildSummaryPrompt 三处对齐; 推断标注(综合推断)、缺真实信息标注待补充交给用户、绝不凭空编 Tests: test_roundtable_seat_modes / test_roundtable_gather_phase / test_roundtable_summary_prompt Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
75 lines
4.5 KiB
Python
75 lines
4.5 KiB
Python
"""ORM model for roundtable-planning business chains.
|
||
|
||
A "business chain" (业务链条) is a named, reusable, *ordered* list of agents the
|
||
user assembles in the chain editor. In Step 2 the coordinator dispatches to the
|
||
seats following this order (see ``multi-agent-business-chain-dev.md``).
|
||
|
||
Persisted per user (like ``roundtable_drafts``) so it follows the account across
|
||
devices/sessions. ``seats`` is stored as a JSON array serialized into
|
||
``PortableLongText`` (MySQL ``LONGTEXT``) rather than ``PortableJSON`` for
|
||
headroom, mirroring the sibling roundtable persistence modules.
|
||
"""
|
||
|
||
from __future__ import annotations
|
||
|
||
from datetime import UTC, datetime
|
||
|
||
from sqlalchemy import Boolean, String
|
||
from sqlalchemy.orm import Mapped, mapped_column
|
||
|
||
from deerflow.persistence.base import Base
|
||
from deerflow.persistence.types import BeijingDateTime, PortableLongText
|
||
|
||
|
||
class RoundtableChainRow(Base):
|
||
__tablename__ = "roundtable_chains"
|
||
|
||
id: Mapped[str] = mapped_column(String(64), primary_key=True)
|
||
user_id: Mapped[str] = mapped_column(String(64), nullable=False, index=True)
|
||
# Human-readable chain title shown on the picker card.
|
||
title: Mapped[str] = mapped_column(String(512), nullable=False, default="")
|
||
# Optional free-text description.
|
||
description: Mapped[str | None] = mapped_column(String(2048), nullable=True)
|
||
# JSON: ordered [{"agent_id": str, "name": str, "description": str}, ...].
|
||
# seats[0] is the first seat to speak. Stored as LONGTEXT-backed JSON string.
|
||
seats: Mapped[str | None] = mapped_column(PortableLongText, nullable=True)
|
||
# Optional JSON: layered orchestration grouping over ``seats`` — a list of
|
||
# stages, each a list of agent_ids run in parallel; stages run serially in
|
||
# order. e.g. [["a","b"],["c"]] = (a‖b) → c. NULL/absent = linear chain
|
||
# (each seat its own stage). Drives Step 2 dag-mode orchestration; see
|
||
# frontend ``lib/dag-types.ts``. Stored as LONGTEXT-backed JSON string.
|
||
stages: Mapped[str | None] = mapped_column(PortableLongText, nullable=True)
|
||
# Optional JSON: per-stage goals aligned by index to ``stages`` — a list where
|
||
# ``stage_goals[i]`` is the「本阶段目标」for ``stages[i]`` (what that serial stage
|
||
# should produce / how it depends on the previous stage's output). Injected to the
|
||
# coordinator dispatch prompt (and seat tasks as fallback) in dag-mode only. NULL =
|
||
# no goals (and always NULL when ``stages`` is NULL/linear). Stored as JSON string.
|
||
stage_goals: Mapped[str | None] = mapped_column(PortableLongText, nullable=True)
|
||
# 可选「总控编排提示」：用户为本链条写给**总控（协调）智能体**的一段指引，告诉总控该
|
||
# 怎么派活、做什么（整体编排策略 / 风格 / 约束）。**仅分层 DAG 模式生效**：注入到总控
|
||
# 每阶段派活提示（``buildDagStageLeaderTask``）。NULL = 不注入。纯文本，存为 LONGTEXT。
|
||
coordinator_prompt: Mapped[str | None] = mapped_column(PortableLongText, nullable=True)
|
||
# 可选「默认席位执行模式」：flash/thinking/pro/ultra（见前端 lib/seat-mode.ts）。选入研讨时
|
||
# 弹窗默认用它（仍可临时改）。NULL = 未设，调用方回退默认(pro)。短串，VARCHAR(16)。
|
||
seat_mode: Mapped[str | None] = mapped_column(String(16), nullable=True)
|
||
# 可选「研讨前并行取数」开关：True = 选入研讨时，研讨前先跑一轮全席位并行『取数』（各席位只调
|
||
# 技能取数、不研讨），产出作为共享资料注入后续研讨轮，把最慢的技能延迟前置并行、给串行研讨段提速。
|
||
# 默认 False（不变更现有行为）。逐条 opt-in。BOOLEAN，server_default "0"。
|
||
gather_first: Mapped[bool] = mapped_column(
|
||
Boolean, nullable=False, default=False, server_default="0"
|
||
)
|
||
# 发布开关：True = 公共（所有账号可在「公共」筛选里查看 / 选用）；False = 仅本人可见。
|
||
# 编辑 / 删除始终只限 owner；发布只是放开「查看 + 选入研讨」。索引以支持「公共」列表筛选。
|
||
is_public: Mapped[bool] = mapped_column(
|
||
Boolean, nullable=False, default=False, server_default="0", index=True
|
||
)
|
||
created_at: Mapped[datetime] = mapped_column(
|
||
BeijingDateTime(), nullable=False, default=lambda: datetime.now(UTC)
|
||
)
|
||
updated_at: Mapped[datetime] = mapped_column(
|
||
BeijingDateTime(),
|
||
nullable=False,
|
||
default=lambda: datetime.now(UTC),
|
||
onupdate=lambda: datetime.now(UTC),
|
||
)
|