Update main.py

main.py

@@ -7,7 +7,8 @@ from flask_cors import CORS
 import random
 import numpy as np
 from langchain_chroma import Chroma
-from chromadb import Documents, EmbeddingFunction, Embeddings
+from chromadb import Documents, EmbeddingFunction, Embeddings, Collection
+import sqlite3
 
 app = Flask(__name__)
 CORS(app)
@@ -73,15 +74,49 @@ class MyEmbeddingFunction(EmbeddingFunction):
             except Exception as e:
                 print("Error in Embeding :",str(e))
 
+# try:
+#     CHROMA_PATH = "chroma"
+#     custom_embeddings = MyEmbeddingFunction()
+#     db = Chroma(
+#          persist_directory=CHROMA_PATH,embedding_function=custom_embeddings
+#     )
+#     #
+# except Exception as e:
+#     print("Error in database :",str(e))
+
+# Initialize the database without persist_directory
 try:
-    CHROMA_PATH = "chroma"
     custom_embeddings = MyEmbeddingFunction()
-    db = Chroma(
-         embedding_function=custom_embeddings
-    )
-    #persist_directory=CHROMA_PATH,
+    db = Chroma(embedding_function=custom_embeddings)
+
+    # Load documents from chroma.sqlite3
+    def load_documents_from_sqlite(db_path="chroma.sqlite3"):
+        conn = sqlite3.connect(db_path)
+        cursor = conn.cursor()
+        
+        # Assuming your table structure has "id", "content", and "embedding"
+        cursor.execute("SELECT id, content, embedding FROM documents")
+        rows = cursor.fetchall()
+        
+        collection = db.get_or_create_collection("default_collection")
+        
+        for row in rows:
+            doc_id = row[0]
+            content = row[1]
+            embedding = json.loads(row[2])  # If embeddings are stored as JSON strings
+            collection.add(
+                ids=[doc_id],
+                documents=[content],
+                embeddings=[embedding]
+            )
+        
+        conn.close()
+        print("Loaded documents into Chroma.")
+    
+    load_documents_from_sqlite()  # Call to load data
+
 except Exception as e:
-    print("Error in database :",str(e))
+    print("Error initializing database:", str(e))
 
 
 def embeddingGen(query):