Refactor backend configuration and enhance news fetching functionality

- Introduced a Config dataclass in config.py to manage API keys, RSS feeds, and directory paths more effectively.
- Updated the NewsFetcher class to include retry logic for fetching articles from RSS feeds.
- Modified the EmbeddingGenerator and NewsRecommender classes to utilize the new configuration structure.
- Enhanced main.py to implement API token verification for secure access to news fetching and recommendations.

backend/vector_store.py

@@ -1,16 +1,11 @@
 from pinecone import Pinecone, ServerlessSpec
-from typing import List, Dict, Any
-from config import (
-    PINECONE_API_KEY,
-    PINECONE_INDEX_NAME,
-    VECTOR_DIMENSION,
-    TOP_K_RESULTS
-)
+from typing import List, Dict, Any, Optional
+from config import config
 
 class VectorStore:
-    def __init__(self):
-        self.pinecone = Pinecone(api_key=PINECONE_API_KEY)
-        self.index_name = PINECONE_INDEX_NAME
+    def __init__(self, pinecone_client: Optional[Pinecone] = None):
+        self.pinecone = pinecone_client or Pinecone(api_key=config.pinecone_api_key)
+        self.index_name = config.pinecone_index_name
         self._ensure_index()
 
     def _ensure_index(self):
@@ -20,11 +15,11 @@ class VectorStore:
             # Create a new index with the correct dimension
             self.pinecone.create_index(
                 name=self.index_name,
-                dimension=VECTOR_DIMENSION,
+                dimension=config.vector_dimension,
                 metric="cosine",
                 spec=ServerlessSpec(cloud="aws", region="us-east-1")
             )
-            print(f"Created new index '{self.index_name}' with dimension {VECTOR_DIMENSION}")
+            print(f"Created new index '{self.index_name}' with dimension {config.vector_dimension}")
         
         self.index = self.pinecone.Index(self.index_name)
 
@@ -57,12 +52,12 @@ class VectorStore:
             print(f"Error upserting articles: {str(e)}")
             return False
 
-    def search_similar(self, query_embedding: List[float], top_k: int = TOP_K_RESULTS) -> List[Dict[str, Any]]:
+    def search_similar(self, query_embedding: List[float], top_k: int = None) -> List[Dict[str, Any]]:
         """Search for similar articles using the query embedding."""
         try:
             results = self.index.query(
                 vector=query_embedding,
-                top_k=top_k,
+                top_k=top_k or config.top_k_results,
                 include_metadata=True
             )