Merge remote-tracking branch 'origin/master' into ji

# Conflicts:
#	offline-backend-20260512/backend/app/gateway/deps.py
This commit is contained in:
吉鑫升 committed 2026-05-18 19:35:16 +08:00
commit 449703a6f0
289 files changed
+80131 -4314

No files matched your search

@@ -0,0 +1,11 @@
from deerflow.persistence.ai_writing_sessions.model import AIWritingSessionRow
from deerflow.persistence.ai_writing_sessions.sql import AIWritingSessionRepository
def make_ai_writing_session_store(session_factory) -> AIWritingSessionRepository | None:
if session_factory is None:
return None
return AIWritingSessionRepository(session_factory)
__all__ = ["AIWritingSessionRow", "AIWritingSessionRepository", "make_ai_writing_session_store"]
@@ -0,0 +1,28 @@
"""ORM model for AI writing sessions."""
from __future__ import annotations
from datetime import UTC, datetime
from sqlalchemy import DateTime, String, Text
from sqlalchemy.orm import Mapped, mapped_column
from deerflow.persistence.base import Base
class AIWritingSessionRow(Base):
__tablename__ = "ai_writing_sessions"
id: Mapped[str] = mapped_column(String(64), primary_key=True)
user_id: Mapped[str | None] = mapped_column(String(64), nullable=True, index=True)
title: Mapped[str] = mapped_column(String(512), nullable=False, default="")
user_intent: Mapped[str | None] = mapped_column(String(512), nullable=True)
status: Mapped[str] = mapped_column(String(32), nullable=False, default="in_progress")
draft_title: Mapped[str | None] = mapped_column(String(512), nullable=True)
draft_markdown: Mapped[str | None] = mapped_column(Text, nullable=True)
completed_interventions: Mapped[str | None] = mapped_column(Text, nullable=True) # JSON array
review_result: Mapped[str | None] = mapped_column(Text, nullable=True) # JSON object
# 完整对话时间线(前端组装的 progressEvents + completedInterventions),用于历史回显
transcript: Mapped[str | None] = mapped_column(Text, nullable=True) # JSON object
created_at: Mapped[datetime] = mapped_column(DateTime(timezone=True), default=lambda: datetime.now(UTC))
updated_at: Mapped[datetime] = mapped_column(DateTime(timezone=True), default=lambda: datetime.now(UTC))
@@ -0,0 +1,162 @@
"""SQLAlchemy-backed AI writing session storage."""
from __future__ import annotations
import json
from datetime import UTC, datetime
from sqlalchemy import select
from sqlalchemy.ext.asyncio import AsyncSession, async_sessionmaker
from deerflow.persistence.ai_writing_sessions.model import AIWritingSessionRow
from deerflow.runtime.user_context import AUTO, _AutoSentinel, resolve_user_id
_UNSET = object()
class AIWritingSessionRepository:
def __init__(self, session_factory: async_sessionmaker[AsyncSession]) -> None:
self._sf = session_factory
@staticmethod
def _row_to_dict(row: AIWritingSessionRow) -> dict:
d = row.to_dict()
for k in ("created_at", "updated_at"):
if isinstance(d.get(k), datetime):
d[k] = d[k].isoformat()
for k in ("completed_interventions", "review_result", "transcript"):
val = d.get(k)
if isinstance(val, str):
try:
d[k] = json.loads(val)
except Exception:
d[k] = None
return d
async def create(
self,
*,
id: str,
title: str,
user_intent: str | None = None,
user_id: str | None | _AutoSentinel = AUTO,
) -> dict:
resolved_user_id = resolve_user_id(user_id, method_name="AIWritingSessionRepository.create")
now = datetime.now(UTC)
row = AIWritingSessionRow(
id=id,
user_id=resolved_user_id,
title=title,
user_intent=user_intent,
status="in_progress",
draft_title=None,
draft_markdown=None,
completed_interventions=None,
review_result=None,
transcript=None,
created_at=now,
updated_at=now,
)
async with self._sf() as session:
session.add(row)
await session.commit()
await session.refresh(row)
return self._row_to_dict(row)
async def update(
self,
session_id: str,
*,
user_id: str | None | _AutoSentinel = AUTO,
title=_UNSET,
status=_UNSET,
draft_title=_UNSET,
draft_markdown=_UNSET,
completed_interventions=_UNSET,
review_result=_UNSET,
transcript=_UNSET,
) -> dict | None:
resolved_user_id = resolve_user_id(user_id, method_name="AIWritingSessionRepository.update")
async with self._sf() as session:
row = await session.get(AIWritingSessionRow, session_id)
if row is None:
return None
if resolved_user_id is not None and row.user_id != resolved_user_id:
return None
if title is not _UNSET:
row.title = title
if status is not _UNSET:
row.status = status
if draft_title is not _UNSET:
row.draft_title = draft_title
if draft_markdown is not _UNSET:
row.draft_markdown = draft_markdown
if completed_interventions is not _UNSET:
row.completed_interventions = (
json.dumps(completed_interventions, ensure_ascii=False)
if completed_interventions is not None
else None
)
if review_result is not _UNSET:
row.review_result = (
json.dumps(review_result, ensure_ascii=False)
if review_result is not None
else None
)
if transcript is not _UNSET:
row.transcript = (
json.dumps(transcript, ensure_ascii=False)
if transcript is not None
else None
)
row.updated_at = datetime.now(UTC)
await session.commit()
await session.refresh(row)
return self._row_to_dict(row)
async def get(
self,
session_id: str,
*,
user_id: str | None | _AutoSentinel = AUTO,
) -> dict | None:
resolved_user_id = resolve_user_id(user_id, method_name="AIWritingSessionRepository.get")
async with self._sf() as session:
row = await session.get(AIWritingSessionRow, session_id)
if row is None:
return None
if resolved_user_id is not None and row.user_id != resolved_user_id:
return None
return self._row_to_dict(row)
async def list(
self,
*,
user_id: str | None | _AutoSentinel = AUTO,
limit: int = 50,
) -> list[dict]:
resolved_user_id = resolve_user_id(user_id, method_name="AIWritingSessionRepository.list")
stmt = select(AIWritingSessionRow)
if resolved_user_id is not None:
stmt = stmt.where(AIWritingSessionRow.user_id == resolved_user_id)
stmt = stmt.order_by(AIWritingSessionRow.updated_at.desc()).limit(limit)
async with self._sf() as session:
result = await session.execute(stmt)
return [self._row_to_dict(r) for r in result.scalars()]
async def delete(
self,
session_id: str,
*,
user_id: str | None | _AutoSentinel = AUTO,
) -> bool:
resolved_user_id = resolve_user_id(user_id, method_name="AIWritingSessionRepository.delete")
async with self._sf() as session:
row = await session.get(AIWritingSessionRow, session_id)
if row is None:
return False
if resolved_user_id is not None and row.user_id != resolved_user_id:
return False
await session.delete(row)
await session.commit()
return True
@@ -0,0 +1,11 @@
from deerflow.persistence.article_types.model import ArticleTypeRow
from deerflow.persistence.article_types.sql import ArticleTypeRepository
def make_article_type_store(session_factory) -> ArticleTypeRepository | None:
if session_factory is None:
return None
return ArticleTypeRepository(session_factory)
__all__ = ["ArticleTypeRow", "ArticleTypeRepository", "make_article_type_store"]
@@ -0,0 +1,24 @@
"""ORM model for AI writing article types."""
from __future__ import annotations
from datetime import UTC, datetime
from sqlalchemy import DateTime, Integer, String, Text
from sqlalchemy.orm import Mapped, mapped_column
from deerflow.persistence.base import Base
class ArticleTypeRow(Base):
__tablename__ = "ai_writing_article_types"
id: Mapped[str] = mapped_column(String(64), primary_key=True)
key: Mapped[str] = mapped_column(String(64), nullable=False, unique=True)
label: Mapped[str] = mapped_column(String(128), nullable=False)
description: Mapped[str | None] = mapped_column(Text, nullable=True)
prompt_hint: Mapped[str | None] = mapped_column(Text, nullable=True)
default_word_count: Mapped[int] = mapped_column(Integer, nullable=False, default=800)
sort_order: Mapped[int] = mapped_column(Integer, nullable=False, default=0)
created_at: Mapped[datetime] = mapped_column(DateTime(timezone=True), default=lambda: datetime.now(UTC))
updated_at: Mapped[datetime] = mapped_column(DateTime(timezone=True), default=lambda: datetime.now(UTC))
@@ -0,0 +1,123 @@
"""SQLAlchemy-backed article type storage."""
from __future__ import annotations
import uuid
from datetime import UTC, datetime
from sqlalchemy import select
from sqlalchemy.ext.asyncio import AsyncSession, async_sessionmaker
from deerflow.persistence.article_types.model import ArticleTypeRow
_UNSET = object()
class ArticleTypeRepository:
def __init__(self, session_factory: async_sessionmaker[AsyncSession]) -> None:
self._sf = session_factory
@staticmethod
def _row_to_dict(row: ArticleTypeRow) -> dict:
d = row.to_dict()
for k in ("created_at", "updated_at"):
if isinstance(d.get(k), datetime):
d[k] = d[k].isoformat()
return d
async def create(
self,
*,
key: str,
label: str,
description: str | None = None,
prompt_hint: str | None = None,
default_word_count: int = 800,
sort_order: int = 0,
) -> dict:
now = datetime.now(UTC)
row = ArticleTypeRow(
id=str(uuid.uuid4()),
key=key,
label=label,
description=description,
prompt_hint=prompt_hint,
default_word_count=default_word_count,
sort_order=sort_order,
created_at=now,
updated_at=now,
)
async with self._sf() as session:
session.add(row)
await session.commit()
await session.refresh(row)
return self._row_to_dict(row)
async def update(
self,
article_type_id: str,
*,
key=_UNSET,
label=_UNSET,
description=_UNSET,
prompt_hint=_UNSET,
default_word_count=_UNSET,
sort_order=_UNSET,
) -> dict | None:
async with self._sf() as session:
row = await session.get(ArticleTypeRow, article_type_id)
if row is None:
return None
if key is not _UNSET:
row.key = key
if label is not _UNSET:
row.label = label
if description is not _UNSET:
row.description = description
if prompt_hint is not _UNSET:
row.prompt_hint = prompt_hint
if default_word_count is not _UNSET:
row.default_word_count = default_word_count
if sort_order is not _UNSET:
row.sort_order = sort_order
row.updated_at = datetime.now(UTC)
await session.commit()
await session.refresh(row)
return self._row_to_dict(row)
async def get(self, article_type_id: str) -> dict | None:
async with self._sf() as session:
row = await session.get(ArticleTypeRow, article_type_id)
if row is None:
return None
return self._row_to_dict(row)
async def get_by_key(self, key: str) -> dict | None:
async with self._sf() as session:
stmt = select(ArticleTypeRow).where(ArticleTypeRow.key == key)
result = await session.execute(stmt)
row = result.scalar_one_or_none()
if row is None:
return None
return self._row_to_dict(row)
async def list(self) -> list[dict]:
stmt = select(ArticleTypeRow).order_by(ArticleTypeRow.sort_order.asc(), ArticleTypeRow.created_at.asc())
async with self._sf() as session:
result = await session.execute(stmt)
return [self._row_to_dict(r) for r in result.scalars()]
async def delete(self, article_type_id: str) -> bool:
async with self._sf() as session:
row = await session.get(ArticleTypeRow, article_type_id)
if row is None:
return False
await session.delete(row)
await session.commit()
return True
async def list_keys(self) -> list[str]:
stmt = select(ArticleTypeRow.key).order_by(ArticleTypeRow.sort_order.asc())
async with self._sf() as session:
result = await session.execute(stmt)
return list(result.scalars())
@@ -0,0 +1,50 @@
"""Create ai_writing_sessions table.
Revision ID: 20260517_01
Revises: 20260515_01
Create Date: 2026-05-17
"""
from __future__ import annotations
import sqlalchemy as sa
from alembic import op
from sqlalchemy import inspect
revision = "20260517_01"
down_revision = "20260515_01"
branch_labels = None
depends_on = None
def _has_table(table_name: str) -> bool:
inspector = inspect(op.get_bind())
return table_name in inspector.get_table_names()
def upgrade() -> None:
if _has_table("ai_writing_sessions"):
return
op.create_table(
"ai_writing_sessions",
sa.Column("id", sa.String(64), primary_key=True),
sa.Column("user_id", sa.String(64), nullable=True),
sa.Column("title", sa.String(512), nullable=False, server_default=""),
sa.Column("user_intent", sa.String(512), nullable=True),
sa.Column("status", sa.String(32), nullable=False, server_default="in_progress"),
sa.Column("draft_title", sa.String(512), nullable=True),
sa.Column("draft_markdown", sa.Text, nullable=True),
sa.Column("completed_interventions", sa.Text, nullable=True),
sa.Column("review_result", sa.Text, nullable=True),
sa.Column("created_at", sa.DateTime(timezone=True), nullable=False),
sa.Column("updated_at", sa.DateTime(timezone=True), nullable=False),
)
op.create_index("ix_ai_writing_sessions_user_id", "ai_writing_sessions", ["user_id"])
op.create_index("ix_ai_writing_sessions_updated_at", "ai_writing_sessions", ["updated_at"])
def downgrade() -> None:
if _has_table("ai_writing_sessions"):
op.drop_table("ai_writing_sessions")
@@ -0,0 +1,107 @@
"""Create ai_writing_article_types table with default types.
Revision ID: 20260517_02
Revises: 20260517_01
Create Date: 2026-05-17
"""
from __future__ import annotations
import uuid
from datetime import UTC, datetime
import sqlalchemy as sa
from alembic import op
from sqlalchemy import inspect
revision = "20260517_02"
down_revision = "20260517_01"
branch_labels = None
depends_on = None
_DEFAULT_TYPES = [
{
"key": "news",
"label": "新闻资讯",
"description": "时效性强的新闻报道,注重事实准确、简洁客观",
"prompt_hint": "以新闻报道的风格写作,注重5W1H要素,语言简洁客观,突出时效性和事实准确性",
"default_word_count": 600,
"sort_order": 0,
},
{
"key": "analysis",
"label": "分析报道",
"description": "深入分析事件背景与影响,提供专业见解",
"prompt_hint": "以分析报道的风格写作,提供深度背景分析、多角度解读和专业见解,逻辑清晰",
"default_word_count": 1200,
"sort_order": 1,
},
{
"key": "feature",
"label": "深度特稿",
"description": "人物或事件的深度叙事,注重细节与故事性",
"prompt_hint": "以深度特稿的风格写作,注重叙事性和细节描写,人物刻画生动,故事性强,情感真实",
"default_word_count": 2000,
"sort_order": 2,
},
{
"key": "report",
"label": "调研报告",
"description": "数据驱动的专题研究报告,结构严谨",
"prompt_hint": "以调研报告的风格写作,数据详实,结构严谨(摘要/背景/发现/结论),语言专业规范",
"default_word_count": 3000,
"sort_order": 3,
},
]
def _has_table(table_name: str) -> bool:
inspector = inspect(op.get_bind())
return table_name in inspector.get_table_names()
def upgrade() -> None:
if not _has_table("ai_writing_article_types"):
op.create_table(
"ai_writing_article_types",
sa.Column("id", sa.String(64), primary_key=True),
sa.Column("key", sa.String(64), nullable=False, unique=True),
sa.Column("label", sa.String(128), nullable=False),
sa.Column("description", sa.Text, nullable=True),
sa.Column("prompt_hint", sa.Text, nullable=True),
sa.Column("default_word_count", sa.Integer, nullable=False, server_default="800"),
sa.Column("sort_order", sa.Integer, nullable=False, server_default="0"),
sa.Column("created_at", sa.DateTime(timezone=True), nullable=False),
sa.Column("updated_at", sa.DateTime(timezone=True), nullable=False),
)
# 预置默认文章类型(幂等,跳过已存在的 key)
conn = op.get_bind()
existing_keys = {row[0] for row in conn.execute(sa.text("SELECT key FROM ai_writing_article_types")).fetchall()}
now = datetime.now(UTC)
for t in _DEFAULT_TYPES:
if t["key"] not in existing_keys:
conn.execute(
sa.text(
"INSERT INTO ai_writing_article_types "
"(id, key, label, description, prompt_hint, default_word_count, sort_order, created_at, updated_at) "
"VALUES (:id, :key, :label, :description, :prompt_hint, :default_word_count, :sort_order, :created_at, :updated_at)"
),
{
"id": str(uuid.uuid4()),
"key": t["key"],
"label": t["label"],
"description": t["description"],
"prompt_hint": t["prompt_hint"],
"default_word_count": t["default_word_count"],
"sort_order": t["sort_order"],
"created_at": now,
"updated_at": now,
},
)
def downgrade() -> None:
if _has_table("ai_writing_article_types"):
op.drop_table("ai_writing_article_types")
@@ -0,0 +1,39 @@
"""Add transcript column to ai_writing_sessions.
存储前端组装的完整对话时间线(progressEvents + completedInterventions),
供历史会话回显出与实时对话一致的富时间线。
Revision ID: 20260517_03
Revises: 20260517_02
Create Date: 2026-05-17
"""
from __future__ import annotations
import sqlalchemy as sa
from alembic import op
from sqlalchemy import inspect
revision = "20260517_03"
down_revision = "20260517_02"
branch_labels = None
depends_on = None
def _has_column(table_name: str, column_name: str) -> bool:
inspector = inspect(op.get_bind())
if table_name not in inspector.get_table_names():
return False
return any(col["name"] == column_name for col in inspector.get_columns(table_name))
def upgrade() -> None:
if _has_column("ai_writing_sessions", "transcript"):
return
op.add_column("ai_writing_sessions", sa.Column("transcript", sa.Text, nullable=True))
def downgrade() -> None:
if _has_column("ai_writing_sessions", "transcript"):
op.drop_column("ai_writing_sessions", "transcript")