Merge branch 'RS-101' into 'master'

[RS-101] - Integrate Astradb into Ecom Review Sense

See merge request !2

.env

-ASTRA_DB_API_ENDPOINT="https://0c52512b-0c65-4e70-ac47-71edf5244a82-us-east-2.apps.astra.datastax.com"
-ASTRA_DB_APPLICATION_TOKEN="AstraCS:EvXpFFafufegdQJvhqlYxmxt:ef86b5996013b12140b69254bd554d7e8e10eb5a7137859b9c432f92a5a3b65c"
-ASTRA_DB_NAMESPACE="default_keyspace"
-HF_TOKEN="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"
-GROQ_API_KEY="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"
+GROQ_API_KEY="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"
+GROQ_MODEL_NAME="llama-3.3-70b-versatile"
 MONGODB_URL=mongodb+srv://genai:nisum@123@productrater.hpbsn.mongodb.net/?retryWrites=true&w=majority&appName=productRater
+MONGODB_PASSWORD=nisum@123
+MONGODB_USERNAME=genai
+MONGODB_CLUSTER=productrater.hpbsn.mongodb.net
+MONGODB_APPNAME=productRater
+
+# Default settings
+EMBEDDING_MODEL="sentence-transformers/all-mpnet-base-v2"
+HUGGING_FACE_INTERFACE_API="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"
+HUGGING_FACE_MODEL="BAAI/bge-base-en-v1.5"
+
+ASTRA_DB_API_END_POINT="https://9706ee0b-b3e8-4ee2-bb26-45e8a0db1586-us-east-2.apps.astra.datastax.com"
+ASTRA_DB_TOKEN="AstraCS:OEbpcggZbUjOvRiesULZkTnf:9d215d8d32c2ffb09081212208e474f81a0bcdde45e7b683cd67f4a1a936a8bf"
+ASTRA_DB_NAME_SPACE="default_keyspace"
+

src/reviewsense_ecom/llm/llm.py

 from functools import lru_cache
 
 from langchain_groq import ChatGroq
+import os
+from dotenv import load_dotenv
 
+load_dotenv()
+groq_model_name = os.getenv("GROQ_MODEL_NAME")
 
 @lru_cache(maxsize=1)  # Cache only one instance (singleton behavior)
 def get_llm(groq_token: str):
@@ -15,7 +19,7 @@ def get_llm(groq_token: str):
         ChatGroq: A singleton instance of the ChatGroq LLM.
     """
     return ChatGroq(
-        model_name="llama-3.3-70b-versatile",
+        model_name=groq_model_name,
         temperature=1,
         groq_api_key=groq_token  # Pass the Groq token here
     )

src/reviewsense_ecom/mongo/mongo_db_config.py

@@ -6,7 +6,20 @@ from pymongo import MongoClient
 from src.reviewsense_ecom.model.FeatureSentiment import FeatureSentiment
 from src.reviewsense_ecom.model.Product import Product
 from src.reviewsense_ecom.model.ProductReview import ProductReview
+from src.reviewsense_ecom.service.review_adder import ReviewAdder
+import os
+from dotenv import load_dotenv
 
+load_dotenv()
+mongo_db = os.getenv("MONGODB_URL")
+# Load environment variables
+load_dotenv()
+
+# Read individual parts from .env
+username = os.getenv("MONGODB_USERNAME")
+password = urllib.parse.quote_plus(os.getenv("MONGODB_PASSWORD"))  # escape special characters
+cluster = os.getenv("MONGODB_CLUSTER")
+appname = os.getenv("MONGODB_APPNAME")
 
 # def get_db_connection(collection_name: str):   #LOCAL DB
 #     client = MongoClient("mongodb://localhost:27017/")  # Update with your MongoDB connection
@@ -15,10 +28,8 @@ from src.reviewsense_ecom.model.ProductReview import ProductReview
 
 
 def get_db_connection(collection_name: str):  # get_db_cloud_connection
-    password = "nisum@123"
-    escaped_password = urllib.parse.quote_plus(password)
-    client = MongoClient(
-        f"mongodb+srv://genai:{escaped_password}@productrater.hpbsn.mongodb.net/?retryWrites=true&w=majority&appName=productRater")  # Update with your MongoDB connection
+    mongo_uri = f"mongodb+srv://{username}:{password}@{cluster}/?retryWrites=true&w=majority&appName={appname}"
+    client = MongoClient(mongo_uri)
     db = client["productRater"]  # Change to your database name
     return db[collection_name]
 
@@ -134,7 +145,14 @@ def add_review_features(input_data, reviews_by_feature):
     inserted_result = collection.insert_one(new_review.__dict__)
 
     if inserted_result.inserted_id:
-        print(f"Review successfully added with ID: {inserted_result.inserted_id}")
+        print(f"Review successfully added Into MongoDB with ID: {inserted_result.inserted_id}")
+    # ✅ Add review to vector DB
+    try:
+        review_adder = ReviewAdder()
+        vector_id = review_adder.add_review_vector(input_data.product_id, input_data.new_review)
+        print(f"Review also added to Vector DB with ID: {vector_id} ")
+    except Exception as e:
+        print(f"Failed to add review to vector DB: {e}")
 
     # Return the newly inserted review object
     return new_review.__dict__

src/reviewsense_ecom/retriever/retrieval.py

 # src/reviewsense/core/database.py
 from functools import lru_cache
+from dotenv import load_dotenv
+import os
 
-from langchain_astradb import AstraDBVectorStore
+from langchain_community.embeddings import HuggingFaceInferenceAPIEmbeddings
 from langchain_huggingface import HuggingFaceEmbeddings
+from langchain_astradb import AstraDBVectorStore
 
+load_dotenv()
+huggingface_inference_api = os.getenv("HUGGING_FACE_INTERFACE_API")
+huggingface_model= os.getenv("HUGGING_FACE_MODEL")
+astra_db_api_end_point=os.getenv("ASTRA_DB_API_END_POINT")
+astra_db_token=os.getenv("ASTRA_DB_TOKEN")
+astra_db_name_space=os.getenv("ASTRA_DB_NAME_SPACE")
 
 @lru_cache(maxsize=1)
 def get_vector_store(
-        collection_name: str = "feature_reviews",
-        embedding_model: str = "sentence-transformers/all-mpnet-base-v2"
+    collection_name: str = "customer_reviews",
+    embedding_model: str = "sentence-transformers/all-mpnet-base-v2"
 ):
     """
     Create a singleton vector store instance
@@ -22,13 +31,13 @@ def get_vector_store(
     """
 
     # Initialize embeddings
-    embeddings = HuggingFaceEmbeddings(model_name=embedding_model)
-
+    embeddings = HuggingFaceInferenceAPIEmbeddings(api_key=huggingface_inference_api, model_name=huggingface_model)
+    
     # Create and return vector store
     return AstraDBVectorStore(
         collection_name=collection_name,
         embedding=embeddings,
-        api_endpoint="https://19b6fcda-3fe8-4585-a5f5-6a464a382426-westus3.apps.astra.datastax.com",
-        token="AstraCS:UaZmElcIDkRHzUmktguKwrnd:197d3e825510580f5f97cc749b00a94b685d54a5f1d4a755b75440fe2c29329b",
-        namespace="default_keyspace",
-    )
+        api_endpoint=astra_db_api_end_point,
+        token=astra_db_token,
+        namespace=astra_db_name_space,
+    )
\ No newline at end of file

src/reviewsense_ecom/service/FeatureExtractor.py

 from typing import List, Dict
 
+import os
 from dotenv import load_dotenv
 from langchain.prompts import ChatPromptTemplate
 from langchain_core.output_parsers import JsonOutputParser
@@ -9,7 +10,7 @@ from src.reviewsense_ecom.llm.llm import get_llm
 
 # Load environment variables from .env file
 load_dotenv()
-
+groq_api_key = os.getenv("GROQ_API_KEY")
 
 class FeatureReview(BaseModel):
     feature: str
@@ -24,7 +25,7 @@ class FeatureReviews(BaseModel):
 
 class FeatureExtractor:
     def __init__(self):
-        self.llm = get_llm("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")
+        self.llm = get_llm(groq_api_key)
         self.parser = self._create_reviews_parser()
         self.prompt = self._create_extraction_prompt()
 

src/reviewsense_ecom/service/review_adder.py

+# services/review_adder.py
+
+from langchain_community.docstore.document import Document
+from src.reviewsense_ecom.retriever.retrieval import get_vector_store
+
+class ReviewAdder:
+    """Class for adding reviews to the vector store"""
+
+    def __init__(self):
+        self.vector_store = get_vector_store()
+
+    def add_review_vector(self, product_id: str, review: str) -> str:
+        review_document = Document(page_content=review, metadata={"title": product_id})
+        return self.vector_store.add_documents([review_document])[0]