Address Copilot feedback

Author: pamelafox

requirements.txt

@@ -6,4 +6,5 @@ azure-search-documents
 python-dotenv
 pillow
 azure-ai-evaluation
-ipykernel
+ipykernel
+rich

zava_product_upload.py

@@ -132,19 +132,10 @@ def create_index(index_client: SearchIndexClient, index_name: str) -> None:
         index_client: Azure Search Index Client
         index_name: Name of the index to create
     """
-    print(f"Checking if index '{index_name}' exists...")
-    
-    existing_indexes = [index.name for index in index_client.list_indexes()]
-    
-    if index_name in existing_indexes:
-        print(f"Index '{index_name}' already exists. Deleting it...")
-        index_client.delete_index(index_name)
-        print(f"Index '{index_name}' deleted.")
-    
-    print(f"Creating index '{index_name}'...")
+    print(f"Creating or updating index '{index_name}'...")
     index_schema = create_product_index_schema(index_name)
-    index_client.create_index(index_schema)
-    print(f"Index '{index_name}' created successfully.")
+    index_client.create_or_update_index(index_schema)
+    print(f"Index '{index_name}' created/updated successfully.")
 
 
 def generate_embeddings(openai_client: OpenAI, products: list[dict[str, Any]]) -> None:

zava_search_keyword.py

@@ -17,7 +17,6 @@
 )
 
 search_query = "25 foot hose"
-search_query = "water plants efficiently without waste"
 
 results = search_client.search(search_text=search_query, top=5)
 

zava_search_reranker.py

@@ -23,17 +23,18 @@
     credential=azure_credential,
 )
 
-search_query = "100 foot hose that wont break"
+search_query = "100 foot hose that won't break"
+
 search_vector = openai_client.embeddings.create(
     model=os.environ["AZURE_OPENAI_EMBEDDING_DEPLOYMENT"],
     input=search_query).data[0].embedding
 
-results = list(search_client.search(
+results = search_client.search(
     search_query,
     top=5,
     vector_queries=[
         VectorizedQuery(vector=search_vector, k_nearest_neighbors=50, fields="embedding")],
     query_type="semantic",
-    semantic_configuration_name="products-semantic-config"))
+    semantic_configuration_name="products-semantic-config")
 
 render_product_results(results, f"Hybrid Search with Reranker Results for '{search_query}'", show_reranker=True)

zava_search_rrf.py

@@ -23,14 +23,13 @@
     credential=azure_credential,
 )
 
-search_query = "100 foot hose that wont break"
+search_query = "100 foot hose that won't break"
 
-#search_query = "water plants efficiently without waste"
 search_vector = openai_client.embeddings.create(
     model=os.environ["AZURE_OPENAI_EMBEDDING_DEPLOYMENT"],
     input=search_query).data[0].embedding
 
-results = list(search_client.search(search_query, top=5, vector_queries=[
-    VectorizedQuery(vector=search_vector, k_nearest_neighbors=50, fields="embedding")]))
+results = search_client.search(search_query, top=5, vector_queries=[
+    VectorizedQuery(vector=search_vector, k_nearest_neighbors=50, fields="embedding")])
 
 render_product_results(results, f"Hybrid RRF search results for '{search_query}'")

zava_search_vector.py

@@ -23,13 +23,13 @@
     credential=azure_credential,
 )
 
-search_query = "water plants efficiently without waste"
-search_query = "100 foot hose that wont break"
+search_query = "100 foot hose that won't break"
+
 search_vector = openai_client.embeddings.create(
     model=os.environ["AZURE_OPENAI_EMBEDDING_DEPLOYMENT"],
     input=search_query).data[0].embedding
 
-results = list(search_client.search(None, top=5, vector_queries=[
-    VectorizedQuery(vector=search_vector, k_nearest_neighbors=50, fields="embedding")]))
+results = search_client.search(None, top=5, vector_queries=[
+    VectorizedQuery(vector=search_vector, k_nearest_neighbors=50, fields="embedding")])
 
 render_product_results(results, f"Vector search results for '{search_query}'")