Update app.py

app.py

@@ -181,69 +181,65 @@ def embedding_search_hf_spaces(query: str = "", limit: int = 3) -> Dict:
         Dictionary containing search results with MCP information
     """
     try:
-        # Initialize Pinecone and OpenAI
+        print("[DEBUG] embedding_search_hf_spaces called")
         pinecone_api_key = os.getenv('PINECONE_API_KEY')
         openai_api_key = os.getenv('OPENAI_API_KEY')
-        
+        print(f"[DEBUG] pinecone_api_key exists: {pinecone_api_key is not None}, openai_api_key exists: {openai_api_key is not None}")
         if not pinecone_api_key or not openai_api_key:
+            print("[ERROR] API keys not found")
             return {
                 "error": "API keys not found",
                 "results": [],
                 "total": 0
             }
-        
-        # Initialize clients
+        print("[DEBUG] Initializing Pinecone and OpenAI clients")
         pc = Pinecone(api_key=pinecone_api_key)
         index = pc.Index("hf-mcp")
         client = OpenAI(api_key=openai_api_key)
-        
-        # Generate embedding using OpenAI
+        print("[DEBUG] Generating embedding with OpenAI")
         response = client.embeddings.create(
             input=query,
             model="text-embedding-3-large"
         )
         query_embedding = response.data[0].embedding
-        
-        # Search in Pinecone using the generated embedding
+        print(f"[DEBUG] Embedding generated: {type(query_embedding)}, len={len(query_embedding)}")
+        print("[DEBUG] Querying Pinecone index")
         results = index.query(
             namespace="",
             vector=query_embedding,
             top_k=limit
         )
-        
-        # Process results and get detailed information
+        print(f"[DEBUG] Pinecone query results: {results}")
         space_results = []
         if not results.matches:
+            print("[DEBUG] No matches found in Pinecone results")
             return {
                 "results": [],
                 "total": 0
             }
-        
         for match in results.matches:
             space_id = match.id
             try:
-                # Remove 'spaces/' prefix if present
                 repo_id = space_id.replace('spaces/', '')
-                
-                # Get space information from HF API
+                print(f"[DEBUG] Fetching space info for repo_id: {repo_id}")
                 space = api.space_info(repo_id)
                 space_info = {
                     "id": space.id,
                     "likes": space.likes,
                     "trending_score": space.trending_score,
                     "source": "huggingface",
-                    "score": match.score  # Add similarity score
+                    "score": match.score
                 }
                 space_results.append(space_info)
             except Exception as e:
+                print(f"[ERROR] Error fetching space info for {space_id}: {str(e)}")
                 continue
-        
         return {
             "results": space_results,
             "total": len(space_results)
         }
-        
     except Exception as e:
+        print(f"[CRITICAL ERROR] in embedding_search_hf_spaces: {str(e)}")
         return {
             "error": str(e),
             "results": [],
@@ -262,7 +258,7 @@ def embedding_search_smithery(query: str = "", limit: int = 3) -> Dict:
         Dictionary containing search results with MCP information
     """
     try:
-        # Initialize Pinecone and OpenAI
+        print("[DEBUG] embedding_search_smithery called")
         from pinecone import Pinecone
         from openai import OpenAI
         import os
@@ -270,42 +266,39 @@ def embedding_search_smithery(query: str = "", limit: int = 3) -> Dict:
         pinecone_api_key = os.getenv('PINECONE_API_KEY')
         openai_api_key = os.getenv('OPENAI_API_KEY')
         smithery_token = os.getenv('SMITHERY_TOKEN')
-        
+        print(f"[DEBUG] pinecone_api_key exists: {pinecone_api_key is not None}, openai_api_key exists: {openai_api_key is not None}, smithery_token exists: {smithery_token is not None}")
         if not pinecone_api_key or not openai_api_key or not smithery_token:
+            print("[ERROR] API keys not found")
             return {
                 "error": "API keys not found",
                 "results": [],
                 "total": 0
             }
-        
-        # Initialize clients
+        print("[DEBUG] Initializing Pinecone and OpenAI clients")
         pc = Pinecone(api_key=pinecone_api_key)
         index = pc.Index("smithery-mcp")
         client = OpenAI(api_key=openai_api_key)
-        
-        # Generate embedding using OpenAI
+        print("[DEBUG] Generating embedding with OpenAI")
         response = client.embeddings.create(
             input=query,
             model="text-embedding-3-large"
         )
         query_embedding = response.data[0].embedding
-        
-        # Search in Pinecone using the generated embedding
+        print(f"[DEBUG] Embedding generated: {type(query_embedding)}, len={len(query_embedding)}")
+        print("[DEBUG] Querying Pinecone index")
         results = index.query(
             namespace="",
             vector=query_embedding,
             top_k=limit
         )
-        
-        # Process results and get detailed information from Smithery
+        print(f"[DEBUG] Pinecone query results: {results}")
         server_results = []
         if not results.matches:
+            print("[DEBUG] No matches found in Pinecone results")
             return {
                 "results": [],
                 "total": 0
             }
-        
-        # Prepare headers for Smithery API
         headers = {
             'Authorization': f'Bearer {smithery_token}'
         }
@@ -313,15 +306,14 @@ def embedding_search_smithery(query: str = "", limit: int = 3) -> Dict:
         for match in results.matches:
             server_id = match.id
             try:
-                # Get server information from Smithery API
+                print(f"[DEBUG] Fetching server info for server_id: {server_id}")
                 response = requests.get(
                     f'https://registry.smithery.ai/servers/{server_id}',
                     headers=headers
                 )
-                
                 if response.status_code != 200:
+                    print(f"[ERROR] Smithery API error for {server_id}: {response.status_code}")
                     continue
-                
                 server = response.json()
                 server_info = {
                     "id": server.get('qualifiedName'),
@@ -329,18 +321,18 @@ def embedding_search_smithery(query: str = "", limit: int = 3) -> Dict:
                     "description": server.get('description'),
                     "likes": server.get('useCount', 0),
                     "source": "smithery",
-                    "score": match.score  # Add similarity score
+                    "score": match.score
                 }
                 server_results.append(server_info)
             except Exception as e:
+                print(f"[ERROR] Error fetching server info for {server_id}: {str(e)}")
                 continue
-        
         return {
             "results": server_results,
             "total": len(server_results)
         }
-        
     except Exception as e:
+        print(f"[CRITICAL ERROR] in embedding_search_smithery: {str(e)}")
         return {
             "error": str(e),
             "results": [],