0.1.11 ai修复与全屏

services/ingest_service/app/core/config.py

@@ -1,25 +1,78 @@
-import os
+from __future__ import annotations
+
 from functools import lru_cache
+from pathlib import Path
+from typing import Tuple, Type
+
 from pydantic import Field
 from pydantic_settings import BaseSettings, SettingsConfigDict
+from pydantic_settings.sources import (
+    DotEnvSettingsSource,
+    EnvSettingsSource,
+    PydanticBaseSettingsSource,
+)
+
+
+def _repo_root() -> Path:
+    # services/ingest_service/app/core/config.py -> .../MNOTE
+    return Path(__file__).resolve().parents[4]
 
 
 class Settings(BaseSettings):
-    """FastAPI 配置，统一读取 .env 或系统环境变量"""
-    model_config = SettingsConfigDict(env_file=".env", env_file_encoding="utf-8", case_sensitive=False, extra="ignore")
+    """Ingest Service 配置（优先读取仓库根目录的 .env.local/.env）。"""
 
+    model_config = SettingsConfigDict(
+        env_file=(
+            str(_repo_root() / ".env.local"),
+            str(_repo_root() / ".env"),
+        ),
+        env_file_encoding="utf-8",
+        case_sensitive=False,
+        extra="ignore",
+    )
+
+    # 服务基础
     app_name: str = Field("AIMNOTE Ingest Service", env="INGEST_APP_NAME")
     environment: str = Field("development", env="INGEST_ENV")
     api_prefix: str = "/api"
-    supabase_url: str = Field("http://127.0.0.1:54321", env="SUPABASE_URL")
-    supabase_key: str = Field("", env="SUPABASE_SERVICE_ROLE_KEY")
+
+    # Supabase（通过 Kong: http://127.0.0.1:18000）
+    supabase_url: str = Field("http://127.0.0.1:18000", env="SUPABASE_URL")
+    supabase_service_role_key: str = Field("", env="SUPABASE_SERVICE_ROLE_KEY")
+
+    # LightRAG（独立服务 7777）
     lightrag_url: str = Field("http://127.0.0.1:7777", env="LIGHTRAG_URL")
     lightrag_api_key: str = Field("", env="LIGHTRAG_API_KEY")
-    ollama_base_url: str = Field("http://127.0.0.1:11434/v1", env="OLLAMA_BASE_URL")
-    ollama_api_key: str = Field("ollama", env="OLLAMA_API_KEY")
-    embeddings_model: str = Field("qwen3-embedding:8b", env="EMBEDDING_MODEL")
-    llm_model: str = Field("qwen3:32b", env="DEFAULT_LLM_MODEL")
-    rerank_model: str = Field("qwen3-reranker-4b", env="RERANK_MODEL")
+
+    # 自动入库（rag_index_sources -> LightRAG）
+    auto_index_enabled: bool = Field(True, env="AUTO_INDEX_ENABLED")
+    auto_index_interval_seconds: int = Field(5, env="AUTO_INDEX_INTERVAL_SECONDS")
+    auto_index_batch_size: int = Field(5, env="AUTO_INDEX_BATCH_SIZE")
+    auto_index_max_attempts: int = Field(6, env="AUTO_INDEX_MAX_ATTEMPTS")      
+
+    # 删除宽限期（用于“可撤销删除”）：例如用户误删后 10 分钟内可恢复，宽限期内不清理 OCR / Storage / LightRAG 资源
+    delete_grace_seconds: int = Field(600, env="DELETE_GRACE_SECONDS")
+
+    # MinerU（用于 OCR / PDF 结构化解析）
+    mineru_enabled: bool = Field(True, env="MINERU_ENABLED")
+    mineru_endpoint: str = Field("", env="MINERU_ENDPOINT")
+
+    @classmethod
+    def settings_customise_sources(
+        cls,
+        settings_cls: Type[BaseSettings],
+        init_settings: PydanticBaseSettingsSource,
+        env_settings: EnvSettingsSource,
+        dotenv_settings: DotEnvSettingsSource,
+        file_secret_settings: PydanticBaseSettingsSource,
+    ) -> Tuple[PydanticBaseSettingsSource, ...]:
+        # 关键点：优先使用 .env.local/.env 覆盖系统环境变量，避免被机器上的全局 env 污染。
+        return (
+            dotenv_settings,
+            env_settings,
+            init_settings,
+            file_secret_settings,
+        )
 
 
 @lru_cache()

services/ingest_service/app/main.py

@@ -1,10 +1,9 @@
-import asyncio
 from fastapi import FastAPI
 
 from app.api.routes import router as api_router
 from app.core.config import get_settings
 from app.core.logging import setup_logging
-from app.services.runtime import scheduler, worker
+from app.services.runtime import auto_indexer, scheduler, worker
 
 settings = get_settings()
 
@@ -17,6 +16,17 @@ app.include_router(api_router, prefix=settings.api_prefix)
 @app.on_event("startup")
 async def _startup() -> None:
     scheduler.start()
+    if getattr(settings, "auto_index_enabled", False):
+        try:
+            await auto_indexer.startup()
+        except Exception:  # noqa: BLE001
+            # 启动阶段不阻塞服务：自动入库失败会在后续 tick 中继续重试/记录
+            pass
+        scheduler.add_interval_job(
+            job_id="auto_index_tick",
+            func=auto_indexer.tick,
+            seconds=int(getattr(settings, "auto_index_interval_seconds", 5)),
+        )
     await worker.start()
 
 

services/ingest_service/app/services/runtime.py

@@ -1,6 +1,9 @@
 from app.core.config import get_settings
+from app.services.auto_indexer import AutoIndexService
 from app.services.job_store import InMemoryJobStore
+from app.services.mineru_client import MinerUClient
 from app.services.scheduler import SchedulerManager
+from app.services.supabase_rest import SupabaseRestAsyncClient
 from app.services.tasks import IngestWorker
 from app.services.webhooks import LightRAGWebhook
 
@@ -11,3 +14,8 @@ scheduler = SchedulerManager()
 webhook = LightRAGWebhook()
 worker = IngestWorker(job_store=job_store, webhook=webhook)
 
+# 自动入库依赖：Supabase + MinerU + LightRAG
+supabase = SupabaseRestAsyncClient()
+mineru = MinerUClient()
+auto_indexer = AutoIndexService(supabase=supabase, webhook=webhook, mineru=mineru)
+

services/ingest_service/app/services/webhooks.py

@@ -1,5 +1,5 @@
 import logging
-from typing import Optional
+from typing import Any, Dict, List, Optional
 
 import httpx
 
@@ -42,3 +42,22 @@ class LightRAGWebhook:
             track_id = data.get("track_id") if isinstance(data, dict) else None
             logger.info("LightRAG 入库请求已提交（file_source=%s, track_id=%s）", file_source, track_id)
             return track_id
+
+    async def delete_documents(self, doc_ids: List[str]) -> Dict[str, Any]:
+        """请求 LightRAG 删除文档（delete_document 为后台任务）。
+
+        返回服务端响应（通常包含 status: deletion_started/busy/not_allowed）。
+        """
+        if not doc_ids:
+            return {"status": "not_allowed", "message": "doc_ids 为空", "doc_id": ""}
+
+        payload = {
+            "doc_ids": doc_ids,
+            "delete_file": False,
+            "delete_llm_cache": False,
+        }
+        async with httpx.AsyncClient(timeout=60.0, headers=self._headers()) as client:
+            resp = await client.request("DELETE", f"{self.base_url}/documents/delete_document", json=payload)
+            resp.raise_for_status()
+            data = resp.json()
+            return data if isinstance(data, dict) else {"raw": data}