chore: 提交本地修改

This commit is contained in:
Administrator committed 2026-06-12 18:09:07 +08:00
1 parent 229830ee89
commit 2862eee33b
19 files changed
+78 -218

No files matched your search

@@ -13,6 +13,7 @@ from __future__ import annotations
import json
import logging
import os
import hashlib
from datetime import datetime
from typing import Any
@@ -20,16 +21,13 @@ import httpx
from fastapi import APIRouter, HTTPException
from pydantic import BaseModel, Field
from deerflow.config import get_app_config
from deerflow.config.agents_config import load_agent_config
logger = logging.getLogger(__name__)
router = APIRouter(prefix="/api/knowledge-ingest", tags=["knowledge-ingest"])
# 入库接口地址,可用环境变量覆盖(如切换 UAT/生产)。
KNOWES_INGEST_URL = os.environ.get(
"KNOWES_INGEST_URL",
"https://ch1.b.uat.4.cn/knowes/api/knowledge-outer/add",
)
_REQUEST_TIMEOUT_SECONDS = 30.0
# 标题字段在目标模板里可能叫「名称」或「标题」,按此顺序逐个尝试。
@@ -101,6 +99,21 @@ def _build_payload(template_name: str, batch_number: str, node_name: str, title_
}
def _build_node_name(title: str, timestamp: str) -> str:
"""Build a stable node name as md5(title) + timestamp + .md."""
title_md5 = hashlib.md5(title.encode("utf-8")).hexdigest()
return f"{title_md5}-{timestamp}.md"
def _resolve_ingest_url() -> str:
"""Resolve the external ingest URL from config.yaml, with env fallback."""
config_url = getattr(get_app_config().knowledge_ingest, "url", "").strip()
return os.environ.get(
"KNOWES_INGEST_URL",
config_url or "https://ch1.b.uat.4.cn/knowes/api/knowledge-outer/add",
)
@router.post("/publish", response_model=KnowledgeIngestPublishResponse, summary="Publish markdown artifact to external knowledge base")
async def publish_to_knowledge_base(request: KnowledgeIngestPublishRequest) -> KnowledgeIngestPublishResponse:
title = request.title.strip()
@@ -121,8 +134,10 @@ async def publish_to_knowledge_base(request: KnowledgeIngestPublishRequest) -> K
now = datetime.now()
# nodeName 是入库记录的唯一标识:md 标题 + 时间戳。两次尝试必须共用同一个
# nodeName,避免第一次实际已入库时第二次产生重复记录。
node_name = f"{title}-{now.strftime('%Y%m%d%H%M%S')}"
timestamp = now.strftime("%Y%m%d%H%M%S")
node_name = _build_node_name(title, timestamp)
batch_number = f"md{now.strftime('%Y%m%d')}"
ingest_url = _resolve_ingest_url()
logger.info(
"[knowledge-ingest] 开始入库: agent=%s template=%s nodeName=%s batchNumber=%s title=%r content_len=%d url=%s",
@@ -132,7 +147,7 @@ async def publish_to_knowledge_base(request: KnowledgeIngestPublishRequest) -> K
batch_number,
title,
len(content),
KNOWES_INGEST_URL,
ingest_url,
)
attempts: list[KnowledgeIngestAttempt] = []
@@ -148,7 +163,7 @@ async def publish_to_knowledge_base(request: KnowledgeIngestPublishRequest) -> K
json.dumps({**payload, "list": [{**payload["list"][0], "data": [{"fieldName": d["fieldName"], "value": d["value"][:200]} for d in payload["list"][0]["data"]]}]}, ensure_ascii=False),
)
try:
response = await client.post(KNOWES_INGEST_URL, json=payload, headers=headers)
response = await client.post(ingest_url, json=payload, headers=headers)
except httpx.HTTPError as exc:
logger.error("[knowledge-ingest] 第 %d 次发送请求异常 (标题字段=%s): %s", len(attempts) + 1, title_field, exc)
attempts.append(KnowledgeIngestAttempt(title_field=title_field, error=str(exc)))
@@ -15,9 +15,10 @@ from deerflow.config.checkpointer_config import CheckpointerConfig, load_checkpo
from deerflow.config.database_config import DatabaseConfig
from deerflow.config.extensions_config import ExtensionsConfig
from deerflow.config.gateway_config import GatewayConfig
from deerflow.config.guardrails_config import GuardrailsConfig, load_guardrails_config_from_dict
from deerflow.config.knowledge_config import KnowledgeConfig
from deerflow.config.log_system_config import LogSystemConfig
from deerflow.config.guardrails_config import GuardrailsConfig, load_guardrails_config_from_dict
from deerflow.config.knowledge_config import KnowledgeConfig
from deerflow.config.knowledge_ingest_config import KnowledgeIngestConfig
from deerflow.config.log_system_config import LogSystemConfig
from deerflow.config.login_config import AuthLoginConfig
from deerflow.config.memory_config import MemoryConfig, load_memory_config_from_dict
from deerflow.config.model_config import ModelConfig
@@ -135,8 +136,9 @@ class AppConfig(BaseModel):
run_events: RunEventsConfig = Field(default_factory=RunEventsConfig, description="Run event storage configuration")
checkpointer: CheckpointerConfig | None = Field(default=None, description="Checkpointer configuration")
stream_bridge: StreamBridgeConfig | None = Field(default=None, description="Stream bridge configuration")
browser_context: BrowserContextConfig = Field(default_factory=BrowserContextConfig, description="Optional browser page capture integration")
knowledge: KnowledgeConfig = Field(default_factory=KnowledgeConfig, description="Product knowledge base (obsidian-wiki compatible vault) configuration")
browser_context: BrowserContextConfig = Field(default_factory=BrowserContextConfig, description="Optional browser page capture integration")
knowledge: KnowledgeConfig = Field(default_factory=KnowledgeConfig, description="Product knowledge base (obsidian-wiki compatible vault) configuration")
knowledge_ingest: KnowledgeIngestConfig = Field(default_factory=KnowledgeIngestConfig, description="External knowledge-ingest endpoint configuration")
@classmethod
def resolve_config_path(cls, config_path: str | None = None) -> Path:
@@ -0,0 +1,12 @@
"""Configuration for the external knowledge-ingest endpoint."""
from pydantic import BaseModel, Field
class KnowledgeIngestConfig(BaseModel):
"""External knowledge-ingest configuration (`config.yaml` -> `knowledge_ingest`)."""
url: str = Field(
default="https://ch1.b.uat.4.cn/knowes/api/knowledge-outer/add",
description="External knowledge-ingest endpoint URL.",
)
@@ -32,7 +32,7 @@ class AgentRow(Base):
# ``is_private``. Empty string = "not assigned" → the listing layer treats
# it as the admin-configured default square. Each agent belongs to at most
# one square (single column), so it can never be published to two squares.
square_id: Mapped[str] = mapped_column(String(64), nullable=False, default="", index=True)
square_id: Mapped[str] = mapped_column(String(64), nullable=False, default="", server_default="", index=True)
# NULL = not featured on the homepage selector. Non-NULL = admin-curated
# featured rank (lower sorts first). Only ``published=True`` agents are
# eligible; the router enforces that invariant on writes. BIGINT so the
@@ -36,7 +36,7 @@ class SkillRow(Base):
# Publish square (发布广场) this skill is filed under once published. Pure
# organizational category, no access control. Empty string = unassigned →
# the listing layer resolves it to the default square. One square per skill.
square_id: Mapped[str] = mapped_column(String(64), nullable=False, default="", index=True)
square_id: Mapped[str] = mapped_column(String(64), nullable=False, default="", server_default="", index=True)
created_at: Mapped[datetime] = mapped_column(DateTime(timezone=True), nullable=False, default=lambda: datetime.now(UTC))
updated_at: Mapped[datetime] = mapped_column(DateTime(timezone=True), nullable=False, default=lambda: datetime.now(UTC), onupdate=lambda: datetime.now(UTC))
+2
View File
@@ -108,6 +108,8 @@ vision:
model_name: deepseek-ai/DeepSeek-OCR
# prompt: 请详细描述这张图片的内容。
# max_tokens: 2048
knowledge_ingest:
url: "https://ch1.b.uat.4.cn/knowes/api/knowledge-outer/add"
# 上传文件识别配置。
# - auto_convert_documents: true 时,上传的 PDF / Word / Excel / PPT 会在服务端自动转换为
# Markdown(.md),agent 通过 read_file / grep 读取其内容 —— 即“识别”文档内容。