Changes on including file features

app.py

@@ -7,18 +7,16 @@ import json
 # Carbon API Key
 CARBON_API_KEY = "a38ee1fe5fef56fc8e1ae2afc881378804bb902882442e1554adae4f82ee23ea"
 customer_id = "Candid"
-data_source_id = 1  # Set your actual data_source_id here
 
+# Define your main function here
 def main():
     st.title('Google Drive Data Connector using Carbon SDK')
     
-    # Authenticate with Carbon API
     st.write('### Authenticate with Carbon API')
     carbon = Carbon(api_key=CARBON_API_KEY, customer_id=customer_id)
     token = carbon.auth.get_access_token()
     carbon = Carbon(access_token=token.access_token)  # authenticated object
 
-    # Connect to Google Drive
     st.write('### Connect to Google Drive')
     get_oauth_url_response = carbon.integrations.get_oauth_url(
         service="GOOGLE_DRIVE",
@@ -34,7 +32,7 @@ def main():
         salesforce_domain="string_example",
         sync_files_on_connection=True,
         set_page_as_boundary=False,
-        data_source_id=Decimal(data_source_id),
+        data_source_id=Decimal(1),
         connecting_new_account=False,
         request_id="b7620173-662c-4ae7-bb61-2e6ffd8619f5",
         use_ocr=False,
@@ -50,49 +48,43 @@ def main():
     )
     
     st.write(f"OAuth URL: {get_oauth_url_response.oauth_url}")
-    st.write("Please use the above URL to connect your Google Drive account.")
     
-    # List files in the connected Google Drive
-    st.write('### List Files in Connected Google Drive')
-    if st.button('List Files'):
-        sync_response = carbon.integrations.sync_data_source_items(data_source_id=Decimal(data_source_id))
-        if sync_response:
-            for item in sync_response['items']:
-                st.write(f"File ID: {item['id']}, File Name: {item['name']}")
-        else:
-            st.write("No files found.")
+    st.write('### Fetch Files and Folders')
+    list_files_response = carbon.data.list()
     
-    # Search in the connected data source
+    st.write('Files and Folders:')
+    for item in list_files_response['data']:
+        st.write(f"Name: {item['name']} - Type: {item['type']}")
+
     st.write('### Search in the Connected Data Source')
-    query = st.text_input("Enter your query:", value="Is there a web version of CNN explainer?")
-    if st.button('Search'):
-        if query:
-            url = "https://api.carbon.ai/embeddings"
-            payload = {
-                "query": query,
-                "k": 2,
-                "file_ids": [],  # Modify to include relevant file IDs if needed
-                "include_all_children": True,
-                "tags": {},
-                "include_tags": True,
-                "include_vectors": True,
-                "include_raw_file": True,
-                "hybrid_search": False,
-                "media_type": "TEXT",
-                "embedding_model": "OPENAI"
-            }
-            headers = {
-                "authorization": f"Bearer {CARBON_API_KEY}",
-                "customer-id": customer_id,
-                "Content-Type": "application/json"
-            }
-            response_search = requests.post(url, json=payload, headers=headers)
-            response_search_chunks = json.loads(response_search.text)
-            
-            for i, doc in enumerate(response_search_chunks['documents']):
-                st.write(f"Document {i+1}")
-                st.write(doc['content'])
-                st.write("-------------------------------------------------")
+    query = st.text_input("Enter your query:", value=" ")
+    if query:
+        url = "https://api.carbon.ai/embeddings"
+        payload = {
+            "query": query,
+            "k": 2,
+            "file_ids": [item['id'] for item in list_files_response['data'] if item['type'] == 'file'],  # Get all file ids
+            "include_all_children": True,
+            "tags": {},
+            "include_tags": True,
+            "include_vectors": True,
+            "include_raw_file": True,
+            "hybrid_search": False,
+            "media_type": "TEXT",
+            "embedding_model": "OPENAI"
+        }
+        headers = {
+            "authorization": f"Bearer {CARBON_API_KEY}",
+            "customer-id": customer_id,
+            "Content-Type": "application/json"
+        }
+        response_search = requests.post(url, json=payload, headers=headers)
+        response_search_chunks = json.loads(response_search.text)
+        
+        for i, doc in enumerate(response_search_chunks['documents']):
+            st.write(f"Document {i+1}")
+            st.write(doc['content'])
+            st.write("-------------------------------------------------")
 
 # Call the main function
 if __name__ == '__main__':