Update app.py

app.py

@@ -16,7 +16,7 @@ import asyncio
 import psutil
 import cachetools
 import hashlib
-from huggingface_hub import InferenceClient, HTTPError
+from huggingface_hub import InferenceClient
 
 app = FastAPI()
 
@@ -158,8 +158,8 @@ async def process_with_llm(filename: str, raw_text: str):
                 structured_data_cache[text_hash] = structured_data
                 logger.info(f"LLM API processing for {filename} with {model}, attempt {attempt+1}, took {time.time() - start_time:.2f} seconds, {log_memory_usage()}")
                 return structured_data
-            except HTTPError as e:
-                if e.status_code == 429:  # Rate limit
+            except Exception as e:
+                if hasattr(e, 'response') and e.response.status_code == 429:  # Rate limit
                     logger.warning(f"Rate limit hit for {filename} with {model}, attempt {attempt+1}: {str(e)}, {log_memory_usage()}")
                     if attempt == 1:
                         break
@@ -167,9 +167,6 @@ async def process_with_llm(filename: str, raw_text: str):
                 else:
                     logger.warning(f"LLM API processing failed for {filename} with {model}, attempt {attempt+1}: {str(e)}, {log_memory_usage()}")
                     break
-            except Exception as e:
-                logger.warning(f"LLM API processing failed for {filename} with {model}, attempt {attempt+1}: {str(e)}, {log_memory_usage()}")
-                break
 
     # If all models fail
     error_msg = "All LLM API models failed. Check model availability, authentication, or rate limits."