Update populate_chroma.py

populate_chroma.py

@@ -1,40 +1,31 @@
-import os
 import json
 import chromadb
 import firebase_admin
 from firebase_admin import credentials, firestore
 from encoder import SentenceEncoder
 
-def initialize_firebase():
-    """Initializes the Firebase connection using Hugging Face secrets."""
-    # Get the JSON credentials from the environment variable
-    creds_json_string = os.getenv("FIREBASE_CREDS_JSON")
-    
-    if not creds_json_string:
-        print("❌ FIREBASE_CREDS_JSON secret not found. Cannot initialize Firebase.")
-        return None
-        
+def initialize_firebase_with_file():
+    """Initializes Firebase using a local serviceAccountKey.json file."""
     try:
-        # Convert the JSON string back into a dictionary
-        creds_dict = json.loads(creds_json_string)
-        cred = credentials.Certificate(creds_dict)
+        # Use the service account key file
+        cred = credentials.Certificate("serviceAccountKey.json")
         
-        # Initialize the app (check if it's already initialized)
         if not firebase_admin._apps:
             firebase_admin.initialize_app(cred)
             
         db = firestore.client()
-        print("✅ Firebase connection initialized successfully.")
+        print("✅ Firebase connection initialized from file.")
         return db
     except Exception as e:
-        print(f"❌ Could not initialize Firebase. Error: {e}")
+        print(f"❌ Could not initialize Firebase from file. Error: {e}")
+        print("   - Make sure 'serviceAccountKey.json' has been uploaded to the terminal.")
         return None
 
 def populate_vector_db():
     """
     Reads internships from Firestore, generates embeddings, and populates ChromaDB.
     """
-    db = initialize_firebase()
+    db = initialize_firebase_with_file()
     if db is None:
         return
 
@@ -43,21 +34,21 @@ def populate_vector_db():
     chroma_client = chromadb.PersistentClient(path="/data/chroma_db")
     collection = chroma_client.get_or_create_collection(name="internships")
 
-    # 2. Clear existing data in ChromaDB
+    # 2. Clear existing data
     if collection.count() > 0:
         print(f"ℹ️ Clearing {collection.count()} existing items from ChromaDB.")
         collection.delete(ids=collection.get()['ids'])
 
-    # 3. Fetch all data from Firestore
+    # 3. Fetch data from Firestore
     print("📚 Reading internship data from Firestore...")
     internships_ref = db.collection('internships').stream()
     internships = [doc.to_dict() for doc in internships_ref]
     
     if not internships:
-        print("❌ No internship data found in Firestore to process.")
+        print("❌ No internship data found in Firestore.")
         return
 
-    # 4. Generate embeddings and prepare data for ChromaDB
+    # 4. Generate embeddings
     print(f"🧠 Generating embeddings for {len(internships)} internships...")
     texts = [f"{i['title']}. {i['description']}. Skills: {', '.join(i['skills'])}" for i in internships]
     embeddings = encoder.encode(texts, show_progress_bar=True).tolist()
@@ -67,8 +58,8 @@ def populate_vector_db():
     for i in internships:
         i['skills'] = json.dumps(i['skills'])
         metadatas.append(i)
-    
-    # 5. Add data to ChromaDB
+        
+    # 5. Add to ChromaDB
     print("➕ Adding data to ChromaDB...")
     collection.add(ids=ids, embeddings=embeddings, metadatas=metadatas)
     print(f"✅ Successfully populated ChromaDB with {collection.count()} items.")