Update file_utils.py

file_utils.py

@@ -44,21 +44,46 @@ def fetch_study_guides(user):
     return results
 
 # === GPT Utility ===
-def call_llm(prompt, system_message="You are a helpful AI tutor."):
-    approx_tokens = len(prompt) // 4
-    if approx_tokens > 7000:
-        prompt = prompt[:28000] + "\n\n[Truncated for token limit]"
+def call_llm(prompt=None, system_message="You are a helpful assistant.", messages=None):
+    """
+    Call GPT-4 with either a single prompt or full message history.
+    Handles token limits and returns a single string response.
+
+    Args:
+        prompt (str): Optional one-turn prompt.
+        system_message (str): Optional system message.
+        messages (list): Optional list of chat messages (for multi-turn chat).
+
+    Returns:
+        str: The assistant's reply.
+    """
+    max_tokens = 8192
+    max_safe_chars = 28000  # Roughly ~7000 tokens
+
+    if messages:
+        # Optional: truncate text-heavy messages
+        for msg in messages:
+            if len(msg["content"]) > max_safe_chars:
+                msg["content"] = msg["content"][:max_safe_chars] + "\n\n[Truncated due to token limit]"
 
-    response = client.chat.completions.create(
-        model="gpt-4",
-        messages=[
+    else:
+        # One-shot prompt mode
+        if prompt and len(prompt) > max_safe_chars:
+            prompt = prompt[:max_safe_chars] + "\n\n[Truncated due to token limit]"
+        messages = [
             {"role": "system", "content": system_message},
             {"role": "user", "content": prompt}
-        ],
+        ]
+
+    response = client.chat.completions.create(
+        model="gpt-4",
+        messages=messages,
         temperature=0.7
     )
+
     return response.choices[0].message.content
 
+
 # === File Parsing ===
 def extract_text_from_file(file):
     ext = os.path.splitext(file.name)[1].lower()