Add logging to show which database (Oracle DB or ChromaDB) is being used for retrieval

Author: jasperan

agentic_rag/OraDBVectorStore.py

@@ -225,6 +225,7 @@ def add_repo_chunks(self, chunks: List[Dict[str, Any]], document_id: str):
 
     def query_pdf_collection(self, query: str, n_results: int = 3) -> List[Dict[str, Any]]:
         """Query the PDF documents collection"""
+        print("🔍 [Oracle DB] Querying PDF Collection")
         # Generate Embeddings
         embeddings = self.encoder.encode(query, batch_size=32, show_progress_bar=True)
         new_vector = array.array("f", embeddings)
@@ -250,10 +251,12 @@ def query_pdf_collection(self, query: str, n_results: int = 3) -> List[Dict[str,
             }
             formatted_results.append(result)
 
+        print(f"🔍 [Oracle DB] Retrieved {len(formatted_results)} chunks from PDF Collection")
         return formatted_results
 
     def query_web_collection(self, query: str, n_results: int = 3) -> List[Dict[str, Any]]:
         """Query the web documents collection"""
+        print("🔍 [Oracle DB] Querying Web Collection")
         # Generate Embeddings
         embeddings = self.encoder.encode(query, batch_size=32, show_progress_bar=True)
         new_vector = array.array("f", embeddings)
@@ -279,10 +282,12 @@ def query_web_collection(self, query: str, n_results: int = 3) -> List[Dict[str,
             }
             formatted_results.append(result)
 
+        print(f"🔍 [Oracle DB] Retrieved {len(formatted_results)} chunks from Web Collection")
         return formatted_results
 
     def query_general_collection(self, query: str, n_results: int = 3) -> List[Dict[str, Any]]:
         """Query the general knowledge collection"""
+        print("🔍 [Oracle DB] Querying General Knowledge Collection")
         # Generate Embeddings
         embeddings = self.encoder.encode(query, batch_size=32, show_progress_bar=True)
         new_vector = array.array("f", embeddings)
@@ -308,10 +313,12 @@ def query_general_collection(self, query: str, n_results: int = 3) -> List[Dict[
             }
             formatted_results.append(result)
 
+        print(f"🔍 [Oracle DB] Retrieved {len(formatted_results)} chunks from General Knowledge Collection")
         return formatted_results
 
     def query_repo_collection(self, query: str, n_results: int = 3) -> List[Dict[str, Any]]:
         """Query the repository documents collection"""
+        print("🔍 [Oracle DB] Querying Repository Collection")
         # Generate Embeddings
         embeddings = self.encoder.encode(query, batch_size=32, show_progress_bar=True)
         new_vector = array.array("f", embeddings)
@@ -337,6 +344,7 @@ def query_repo_collection(self, query: str, n_results: int = 3) -> List[Dict[str
             }
             formatted_results.append(result)
 
+        print(f"🔍 [Oracle DB] Retrieved {len(formatted_results)} chunks from Repository Collection")
         return formatted_results
 
     def get_collection_count(self, collection_name: str) -> int:

agentic_rag/local_rag_agent.py

@@ -292,10 +292,16 @@ def _process_query_with_cot(self, query: str) -> Dict[str, Any]:
         try:
             # Get context based on collection type
             if self.collection == "PDF Collection":
+                db_type = "Oracle DB" if self.use_oracle_db else "ChromaDB"
+                print(f"🔄 Using {db_type} for retrieving PDF Collection context")
                 context = self.vector_store.query_pdf_collection(query)
             elif self.collection == "Repository Collection":
+                db_type = "Oracle DB" if self.use_oracle_db else "ChromaDB"
+                print(f"🔄 Using {db_type} for retrieving Repository Collection context")
                 context = self.vector_store.query_repo_collection(query)
             elif self.collection == "Web Knowledge Base":
+                db_type = "Oracle DB" if self.use_oracle_db else "ChromaDB"
+                print(f"🔄 Using {db_type} for retrieving Web Knowledge Base context")
                 context = self.vector_store.query_web_collection(query)
             else:
                 context = []
@@ -389,10 +395,16 @@ def _process_query_standard(self, query: str) -> Dict[str, Any]:
         try:
             # Get context based on collection type
             if self.collection == "PDF Collection":
+                db_type = "Oracle DB" if self.use_oracle_db else "ChromaDB"
+                print(f"🔄 Using {db_type} for retrieving PDF Collection context")
                 context = self.vector_store.query_pdf_collection(query)
             elif self.collection == "Repository Collection":
+                db_type = "Oracle DB" if self.use_oracle_db else "ChromaDB"
+                print(f"🔄 Using {db_type} for retrieving Repository Collection context")
                 context = self.vector_store.query_repo_collection(query)
             elif self.collection == "Web Knowledge Base":
+                db_type = "Oracle DB" if self.use_oracle_db else "ChromaDB"
+                print(f"🔄 Using {db_type} for retrieving Web Knowledge Base context")
                 context = self.vector_store.query_web_collection(query)
             else:
                 context = []

agentic_rag/store.py

@@ -117,6 +117,7 @@ def add_repo_chunks(self, chunks: List[Dict[str, Any]], document_id: str):
 
     def query_pdf_collection(self, query: str, n_results: int = 3) -> List[Dict[str, Any]]:
         """Query the PDF documents collection"""
+        print("📊 [ChromaDB] Querying PDF Collection")
         results = self.pdf_collection.query(
             query_texts=[query],
             n_results=n_results
@@ -131,10 +132,12 @@ def query_pdf_collection(self, query: str, n_results: int = 3) -> List[Dict[str,
             }
             formatted_results.append(result)
 
+        print(f"📊 [ChromaDB] Retrieved {len(formatted_results)} chunks from PDF Collection")
         return formatted_results
 
     def query_web_collection(self, query: str, n_results: int = 3) -> List[Dict[str, Any]]:
         """Query the web documents collection"""
+        print("📊 [ChromaDB] Querying Web Collection")
         results = self.web_collection.query(
             query_texts=[query],
             n_results=n_results
@@ -149,10 +152,12 @@ def query_web_collection(self, query: str, n_results: int = 3) -> List[Dict[str,
             }
             formatted_results.append(result)
 
+        print(f"📊 [ChromaDB] Retrieved {len(formatted_results)} chunks from Web Collection")
         return formatted_results
 
     def query_general_collection(self, query: str, n_results: int = 3) -> List[Dict[str, Any]]:
         """Query the general knowledge collection"""
+        print("📊 [ChromaDB] Querying General Knowledge Collection")
         results = self.general_collection.query(
             query_texts=[query],
             n_results=n_results
@@ -167,10 +172,12 @@ def query_general_collection(self, query: str, n_results: int = 3) -> List[Dict[
             }
             formatted_results.append(result)
 
+        print(f"📊 [ChromaDB] Retrieved {len(formatted_results)} chunks from General Knowledge Collection")
         return formatted_results
 
     def query_repo_collection(self, query: str, n_results: int = 3) -> List[Dict[str, Any]]:
         """Query the repository documents collection"""
+        print("📊 [ChromaDB] Querying Repository Collection")
         results = self.repo_collection.query(
             query_texts=[query],
             n_results=n_results
@@ -185,6 +192,7 @@ def query_repo_collection(self, query: str, n_results: int = 3) -> List[Dict[str
             }
             formatted_results.append(result)
 
+        print(f"📊 [ChromaDB] Retrieved {len(formatted_results)} chunks from Repository Collection")
         return formatted_results
 
 def main():