⚙️ fix: dependency crash, conditional imports for heavier packages (#79)

Author: danny-avila

config.py

@@ -1,15 +1,11 @@
 # config.py
 import os
 import json
-import logging
 import boto3
+import logging
 from enum import Enum
 from datetime import datetime
 from dotenv import find_dotenv, load_dotenv
-from langchain_ollama import OllamaEmbeddings
-from langchain_huggingface import HuggingFaceEmbeddings, HuggingFaceEndpointEmbeddings
-from langchain_aws import BedrockEmbeddings
-from langchain_openai import AzureOpenAIEmbeddings, OpenAIEmbeddings
 from starlette.middleware.base import BaseHTTPMiddleware
 from store_factory import get_vector_store
 
@@ -179,28 +175,40 @@ async def dispatch(self, request, call_next):
 
 def init_embeddings(provider, model):
     if provider == EmbeddingsProvider.OPENAI:
+        from langchain_openai import OpenAIEmbeddings
+
         return OpenAIEmbeddings(
             model=model,
             api_key=RAG_OPENAI_API_KEY,
             openai_api_base=RAG_OPENAI_BASEURL,
             openai_proxy=RAG_OPENAI_PROXY,
         )
     elif provider == EmbeddingsProvider.AZURE:
+        from langchain_openai import AzureOpenAIEmbeddings
+
         return AzureOpenAIEmbeddings(
             azure_deployment=model,
             api_key=RAG_AZURE_OPENAI_API_KEY,
             azure_endpoint=RAG_AZURE_OPENAI_ENDPOINT,
             api_version=RAG_AZURE_OPENAI_API_VERSION,
         )
     elif provider == EmbeddingsProvider.HUGGINGFACE:
+        from langchain_huggingface import HuggingFaceEmbeddings
+
         return HuggingFaceEmbeddings(
             model_name=model, encode_kwargs={"normalize_embeddings": True}
         )
     elif provider == EmbeddingsProvider.HUGGINGFACETEI:
+        from langchain_huggingface import HuggingFaceEndpointEmbeddings
+
         return HuggingFaceEndpointEmbeddings(model=model)
     elif provider == EmbeddingsProvider.OLLAMA:
+        from langchain_ollama import OllamaEmbeddings
+
         return OllamaEmbeddings(model=model, base_url=OLLAMA_BASE_URL)
     elif provider == EmbeddingsProvider.BEDROCK:
+        from langchain_aws import BedrockEmbeddings
+
         session = boto3.Session(
             aws_access_key_id=AWS_ACCESS_KEY_ID,
             aws_secret_access_key=AWS_SECRET_ACCESS_KEY,
@@ -237,9 +245,7 @@ def init_embeddings(provider, model):
     EMBEDDINGS_MODEL = get_env_variable(
         "EMBEDDINGS_MODEL", "amazon.titan-embed-text-v1"
     )
-    AWS_DEFAULT_REGION = get_env_variable(
-        "AWS_DEFAULT_REGION", "us-east-1"
-    )
+    AWS_DEFAULT_REGION = get_env_variable("AWS_DEFAULT_REGION", "us-east-1")
 else:
     raise ValueError(f"Unsupported embeddings provider: {EMBEDDINGS_PROVIDER}")
 
@@ -258,7 +264,9 @@ def init_embeddings(provider, model):
 elif VECTOR_DB_TYPE == VectorDBType.ATLAS_MONGO:
     # Backward compatability check
     if MONGO_VECTOR_COLLECTION:
-        logger.info(f"DEPRECATED: Please remove env var MONGO_VECTOR_COLLECTION and instead use COLLECTION_NAME and ATLAS_SEARCH_INDEX. You can set both as same, but not neccessary. See README for more information.")
+        logger.info(
+            f"DEPRECATED: Please remove env var MONGO_VECTOR_COLLECTION and instead use COLLECTION_NAME and ATLAS_SEARCH_INDEX. You can set both as same, but not neccessary. See README for more information."
+        )
         ATLAS_SEARCH_INDEX = MONGO_VECTOR_COLLECTION
         COLLECTION_NAME = MONGO_VECTOR_COLLECTION
     vector_store = get_vector_store(

requirements.lite.txt

@@ -1,15 +1,15 @@
-langchain==0.1.12
-langchain_community==0.0.34
-langchain_openai==0.0.8
-langchain_core==0.1.45
+langchain==0.3
+langchain_community==0.3
+langchain_openai==0.2.0
+langchain_core==0.3.5
 sqlalchemy==2.0.28
 python-dotenv==1.0.1
 fastapi==0.110.0
 psycopg2-binary==2.9.9
 pgvector==0.2.5
 uvicorn==0.28.0
 pypdf==4.1.0
-unstructured==0.12.6
+unstructured==0.15.13
 markdown==3.6
 networkx==3.2.1
 pandas==2.2.1
@@ -20,11 +20,13 @@ PyJWT==2.8.0
 asyncpg==0.29.0
 python-multipart==0.0.9
 aiofiles==23.2.1
-rapidocr-onnxruntime==1.3.17
+rapidocr-onnxruntime==1.3.24
 opencv-python-headless==4.9.0.80
 pymongo==4.6.3
-langchain-mongodb==0.1.3
+langchain-mongodb==0.2.0
 cryptography==42.0.7
 python-magic==0.4.27
 python-pptx==0.6.23
-xlrd==2.0.1
+xlrd==2.0.1
+langchain-aws==0.2.1
+boto3==1.34.144