Update app.py

app.py

@@ -4,6 +4,8 @@ from dotenv import load_dotenv
 import json
 from deep_translator import GoogleTranslator
 import google.generativeai as genai
+import time
+import random
 
 load_dotenv()
 
@@ -18,36 +20,42 @@ def make_call(data):
     language = items['lang']
     query = items['text']
     query = query.lower()
-    translated = None  # Initialize translated to None, it will hold the final output message
-
-    # Use the GenerativeModel class to interact with Gemini
-    model = genai.GenerativeModel('gemini-pro')  # You can choose a different Gemini model
-
-    try:
-        # Craft your prompt specific for Gemini and desired output format
-        prompt_query = (
-            f"Answer this query in a short message with wisdom, love, and compassion, "
-            f"in context to Bhagavad Gita, that feels like chatting to a person and "
-            f"provide references of shlokas from chapters of Bhagavad Gita which are "
-            f"relevant to the query. Keep the answer short, precise, and simple. "
-            f"Query: {query}"
-        )
-
-        # Generate content using Gemini
-        response = model.generate_content(prompt_query)
-
-        # Access the generated text from the response
-        answer = response.text  # The generated content is in the 'text' attribute
-
-        # Translate the answer
-        translated = GoogleTranslator(source='auto', target=language).translate(answer)
-    except Exception as e:
-        print(f"API call failed for: {e}")
-        # Provide an informative message to the user in case of an error
-        translated = f"An error occurred while fetching the answer: {e}" 
-        # Optionally, you might want to log the full traceback for debugging:
-        # import traceback
-        # traceback.print_exc()
+    translated = None
+
+    model = genai.GenerativeModel('gemini-pro')
+
+    retries = 0
+    max_retries = 5  # You can adjust this number
+    base_delay = 1  # Initial delay in seconds
+
+    while retries < max_retries:
+        try:
+            prompt_query = (
+                f"Answer this query in a short message with wisdom, love, and compassion, "
+                f"in context to Bhagavad Gita, that feels like chatting to a person and "
+                f"provide references of shlokas from chapters of Bhagavad Gita which are "
+                f"relevant to the query. Keep the answer short, precise, and simple. "
+                f"Query: {query}"
+            )
+
+            response = model.generate_content(prompt_query)
+            answer = response.text
+            translated = GoogleTranslator(source='auto', target=language).translate(answer)
+            break  # Exit the loop if the call is successful
+        except Exception as e:
+            if "429 Quota exceeded" in str(e):
+                delay = base_delay * (2 ** retries) + random.uniform(0, 1)  # Exponential backoff with jitter
+                print(f"Quota exceeded. Retrying in {delay:.2f} seconds... (Attempt {retries + 1}/{max_retries})")
+                time.sleep(delay)
+                retries += 1
+            else:
+                print(f"API call failed: {e}")
+                translated = f"An error occurred while fetching the answer: {e}"
+                break  # Exit the loop for other errors
+    else:
+        # This block executes if the loop completes without a successful break (i.e., max_retries reached)
+        translated = "Maximum retry attempts reached. Please try again later."
+
 
     respo = {
         "message": translated,