修改配置

rag_indexer/__init__.py

@@ -26,8 +26,22 @@ Offline RAG Indexer module.
 from .index_builder import IndexBuilder, IndexBuilderConfig, DocstoreConfig
 from .loaders import DocumentLoader
 from .splitters import SplitterType, get_splitter
+from .config import (
+    QDRANT_URL,
+    QDRANT_API_KEY,
+    LLAMACPP_EMBEDDING_URL,
+    LLAMACPP_API_KEY,
+    DB_URI,
+    DOCSTORE_URI,
+    RAG_OCR_LANGUAGES,
+    RAG_DOC_LANGUAGES,
+)
 
 # 从 rag_core 重新导出常用组件
+import sys
+from pathlib import Path
+sys.path.insert(0, str(Path(__file__).parent.parent / "backend"))
+
 from rag_core import (
     LlamaCppEmbedder,
     QdrantVectorStore,
@@ -39,7 +53,7 @@ __version__ = "2.0.0"
 
 __all__ = [
     # 核心构建器与配置
-    "index_builder",
+    "IndexBuilder",
     "IndexBuilderConfig",
     "DocstoreConfig",
     
@@ -50,6 +64,16 @@ __all__ = [
     "SplitterType",
     "get_splitter",
     
+    # 配置
+    "QDRANT_URL",
+    "QDRANT_API_KEY",
+    "LLAMACPP_EMBEDDING_URL",
+    "LLAMACPP_API_KEY",
+    "DB_URI",
+    "DOCSTORE_URI",
+    "RAG_OCR_LANGUAGES",
+    "RAG_DOC_LANGUAGES",
+    
     # 嵌入与向量存储
     "LlamaCppEmbedder",
     "QdrantVectorStore",

rag_indexer/cli.py

@@ -6,13 +6,24 @@ import asyncio
 import logging
 import sys
 from pathlib import Path
+from dotenv import load_dotenv
+
+# 加载 .env 文件
+load_dotenv()
 
 # 添加项目根目录和 backend 目录到 Python 路径
 sys.path.insert(0, str(Path(__file__).parent.parent))
 sys.path.insert(0, str(Path(__file__).parent.parent / "backend"))
 
-from .index_builder import IndexBuilder, IndexBuilderConfig
-from .splitters import SplitterType
+# 导入方式：条件导入，支持作为脚本运行和作为包导入
+if __name__ == "__main__":
+    # 作为脚本直接运行时使用绝对导入
+    from rag_indexer.index_builder import IndexBuilder, IndexBuilderConfig
+    from rag_indexer.splitters import SplitterType
+else:
+    # 作为包导入时使用相对导入
+    from .index_builder import IndexBuilder, IndexBuilderConfig
+    from .splitters import SplitterType
 
 logging.basicConfig(
     level=logging.INFO,

rag_indexer/config.py

@@ -1,32 +1,71 @@
 """
 RAG Indexer 配置管理模块
 集中管理所有环境变量配置项，避免散落在各个文件中
+所有配置直接从环境变量读取，无默认值，避免配置混乱
+需要类型转换的配置在此处理
 """
 
 import os
 
-# 尝试从 rag_core 导入配置（如果可用）
-try:
-    from rag_core.config import (
-        QDRANT_URL,
-        QDRANT_API_KEY,
-        LLAMACPP_EMBEDDING_URL,
-        LLAMACPP_API_KEY,
-        DB_URI,
-        DOCSTORE_URI,
-    )
-except ImportError:
-    # 如果 rag_core 不可用，则直接读取环境变量
-    QDRANT_URL = os.getenv("QDRANT_URL", "http://127.0.0.1:6333")
-    QDRANT_API_KEY = os.getenv("QDRANT_API_KEY", "")
-    LLAMACPP_EMBEDDING_URL = os.getenv("LLAMACPP_EMBEDDING_URL", "http://127.0.0.1:8082")
-    LLAMACPP_API_KEY = os.getenv("LLAMACPP_API_KEY", "")
-    DB_URI = os.getenv(
-        "DB_URI",
-        "postgresql://postgres:huang1998@ai-postgres:5432/langgraph_db?sslmode=disable"
-    )
-    DOCSTORE_URI = os.getenv("DOCSTORE_URI", DB_URI)
+
+# ========== 辅助函数：类型转换 ==========
+def _get_str(key: str) -> str | None:
+    """获取字符串配置"""
+    return os.getenv(key)
+
+
+def _get_int(key: str) -> int | None:
+    """获取整数配置，自动转换"""
+    value = os.getenv(key)
+    if value is not None:
+        try:
+            return int(value)
+        except (ValueError, TypeError):
+            pass
+    return None
+
+
+def _get_list_str(key: str, default: list[str] | None = None) -> list[str]:
+    """获取字符串列表配置，从逗号分隔的字符串解析"""
+    value = os.getenv(key)
+    if value is not None:
+        return [item.strip() for item in value.split(",") if item.strip()]
+    return default or []
+
+
+# ========== 向量数据库配置（URL + API密钥 配对） ==========
+QDRANT_URL = _get_str("QDRANT_URL")
+QDRANT_API_KEY = _get_str("QDRANT_API_KEY")
+
+
+# ========== 嵌入服务配置（URL + API密钥 配对） ==========
+LLAMACPP_EMBEDDING_URL = _get_str("LLAMACPP_EMBEDDING_URL")
+LLAMACPP_API_KEY = _get_str("LLAMACPP_API_KEY")
+
+
+# ========== 文档存储配置（分离配置 + 完整URI） ==========
+# 分离配置（优先使用）
+DB_HOST = _get_str("DB_HOST")
+DB_PORT = _get_int("DB_PORT")
+DB_USER = _get_str("DB_USER")
+DB_PASSWORD = _get_str("DB_PASSWORD")
+DB_NAME = _get_str("DB_NAME")
+
+# 完整连接字符串（直接从环境变量读取）
+DB_URI = _get_str("DB_URI")
+
+# 文档存储 URI（直接从环境变量读取，默认同 DB_URI）
+DOCSTORE_URI = _get_str("DOCSTORE_URI") or DB_URI
+
+
+# ========== 文档加载器配置（unstructured 库） ==========
+# OCR 语言列表（逗号分隔，如 "chi_sim,eng"）
+RAG_OCR_LANGUAGES = _get_list_str("RAG_OCR_LANGUAGES", ["chi_sim", "eng"])
+
+# 文档主语言列表（逗号分隔，如 "zh"）
+RAG_DOC_LANGUAGES = _get_list_str("RAG_DOC_LANGUAGES", ["zh"])
+
 
 # ========== 索引器专用配置 ==========
 # 默认索引存储路径
-INDEX_STORAGE_PATH = os.getenv("INDEX_STORAGE_PATH", "./index_storage")
+INDEX_STORAGE_PATH = _get_str("INDEX_STORAGE_PATH")

rag_indexer/index_builder.py

@@ -23,6 +23,12 @@ from qdrant_client.http.exceptions import ResponseHandlingException
 
 from .loaders import DocumentLoader
 from .splitters import SplitterType, get_splitter
+
+# 从 rag_core 导入
+import sys
+from pathlib import Path
+sys.path.insert(0, str(Path(__file__).parent.parent / "backend"))
+
 from rag_core import LlamaCppEmbedder, QdrantVectorStore, create_docstore, create_parent_retriever
 
 logger = logging.getLogger(__name__)

rag_indexer/loaders.py

@@ -11,6 +11,9 @@ from langchain_core.documents import Document
 from unstructured.documents.elements import Element
 from unstructured.partition.auto import partition
 
+# 相对导入配置
+from .config import RAG_OCR_LANGUAGES, RAG_DOC_LANGUAGES
+
 logger = logging.getLogger(__name__)
 
 # 模块加载时设置一次环境变量，避免重复设置
@@ -47,8 +50,8 @@ class DocumentLoader:
         """
         self.extract_images = extract_images
         self.strategy = strategy
-        self.ocr_languages = ocr_languages or ["chi_sim", "eng"]
-        self.languages = languages or ["zh"]
+        self.ocr_languages = ocr_languages or RAG_OCR_LANGUAGES
+        self.languages = languages or RAG_DOC_LANGUAGES
         self.include_page_breaks = include_page_breaks
         self.pdf_infer_table_structure = pdf_infer_table_structure
         self.partition_kwargs = partition_kwargs or {}