spiderforce4ai 2.4.5__py3-none-any.whl → 2.4.7__py3-none-any.whl

spiderforce4ai/post_extraction_agent.py

@@ -12,6 +12,9 @@ from pydantic import BaseModel, Field
 import logging
 from datetime import datetime
 import re
+from rich.console import Console
+
+console = Console()
 
 logger = logging.getLogger(__name__)
 
@@ -190,17 +193,45 @@ class PostExtractionAgent:
                         api_base=self.config.base_url
                     )
                     
-                    # Parse response
-                    extracted_data = json.loads(response.choices[0].message.content)
-                    self.buffer.remove_request(url)  # Remove from buffer if successful
-                    return extracted_data
+                    # Log raw response for debugging
+                    raw_content = response.choices[0].message.content
+                    logger.debug(f"Raw LLM response for {url}: {raw_content}")
                     
-                except json.JSONDecodeError as e:
-                    last_error = f"Invalid JSON response from LLM: {e}"
-                    if attempt < max_retries - 1:
-                        time.sleep(retry_delay * (attempt + 1))
+                    try:
+                        # First try direct JSON parsing
+                        try:
+                            extracted_data = json.loads(raw_content)
+                        except json.JSONDecodeError:
+                            # Look for JSON in markdown code blocks
+                            json_match = re.search(r'```(?:json)?\s*\n([\s\S]*?)\n```', raw_content)
+                            if json_match:
+                                json_content = json_match.group(1).strip()
+                                extracted_data = json.loads(json_content)
+                            else:
+                                # If no JSON found, try to extract structured data in any format
+                                extracted_data = {
+                                    "raw_content": raw_content,
+                                    "format": "text",
+                                    "timestamp": datetime.now().isoformat()
+                                }
+                        
+                        self.buffer.remove_request(url)  # Remove from buffer if successful
+                        return extracted_data
+                    except Exception as e:
+                        error_msg = (
+                            f"Error processing LLM response for {url}:\n"
+                            f"Error: {str(e)}\n"
+                            f"Raw content: {raw_content[:500]}..."  # First 500 chars of response
+                        )
+                        logger.error(error_msg)
+                        last_error = error_msg
+                        if attempt < max_retries - 1:
+                            time.sleep(retry_delay * (attempt + 1))
+                            
                 except Exception as e:
-                    last_error = str(e)
+                    error_msg = f"LLM processing error for {url}: {str(e)}"
+                    logger.error(error_msg)
+                    last_error = error_msg
                     if attempt < max_retries - 1:
                         time.sleep(retry_delay * (attempt + 1))
             
@@ -242,24 +273,42 @@ class PostExtractionAgent:
 
     def process_content(self, url: str, content: str) -> Optional[Dict]:
         """Process content with retry mechanism."""
+        logger.info(f"Starting content processing for {url}")
+        
         for attempt in range(self.config.max_retries):
+            logger.info(f"Processing attempt {attempt + 1}/{self.config.max_retries} for {url}")
+            
             result = self._process_single_content(url, content)
             if result:
+                logger.info(f"Successfully processed content for {url}")
+                
                 # Apply custom transformation if provided
                 if self.config.custom_transform_function:
                     try:
                         result = self.config.custom_transform_function(result)
+                        logger.info(f"Applied custom transformation for {url}")
                     except Exception as e:
-                        logger.error(f"Error in custom transform for {url}: {str(e)}")
+                        error_msg = f"Error in custom transform for {url}: {str(e)}"
+                        logger.error(error_msg)
+                        console.print(f"[red]{error_msg}[/red]")
                 
                 # Save result synchronously
-                self._save_result_sync(url, result)
+                try:
+                    self._save_result_sync(url, result)
+                    logger.info(f"Saved results for {url}")
+                except Exception as e:
+                    error_msg = f"Error saving results for {url}: {str(e)}"
+                    logger.error(error_msg)
+                    console.print(f"[red]{error_msg}[/red]")
+                
                 return result
             
             # Wait before retry
             if attempt < self.config.max_retries - 1:
+                logger.info(f"Attempt {attempt + 1} failed for {url}, waiting {self.config.retry_delay}s before retry")
                 time.sleep(self.config.retry_delay)
         
+        logger.error(f"All processing attempts failed for {url}")
         return None
 
     async def process_bulk_content(self, content_map: Dict[str, str]) -> Dict[str, Optional[Dict]]:

{spiderforce4ai-2.4.5.dist-info → spiderforce4ai-2.4.7.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: spiderforce4ai
-Version: 2.4.5
+Version: 2.4.7
 Summary: Python wrapper for SpiderForce4AI HTML-to-Markdown conversion service with LLM post-processing
 Home-page: https://petertam.pro
 Author: Piotr Tamulewicz

spiderforce4ai-2.4.7.dist-info/RECORD

@@ -0,0 +1,7 @@
+spiderforce4ai/__init__.py,sha256=PPpJLowJhgoRijsF2ebmdkFbIriI_yIFlCi1wL6hSP8,42267
+spiderforce4ai/post_extraction_agent.py,sha256=q2ohsqw_F1e5rT2H9eSzCWzstJLbwGyCtwLsC6eMufs,14560
+spiderforce4ai-2.4.7.dist-info/METADATA,sha256=r273h2ogI76aXTd8XN9b81EWtQLuhdSjZkXD2Ks8GnM,9012
+spiderforce4ai-2.4.7.dist-info/WHEEL,sha256=In9FTNxeP60KnTkGw7wk6mJPYd_dQSjEZmXdBdMCI-8,91
+spiderforce4ai-2.4.7.dist-info/entry_points.txt,sha256=ibARQxOlDiL1ho12zbDZt4Uq5RKSIk_qk159ZlZ46hc,59
+spiderforce4ai-2.4.7.dist-info/top_level.txt,sha256=Kth7A21Js7DCp0j5XBBi-FE45SCLouZkeNZU__Yr9Yk,15
+spiderforce4ai-2.4.7.dist-info/RECORD,,

spiderforce4ai-2.4.5.dist-info/RECORD

@@ -1,7 +0,0 @@
-spiderforce4ai/__init__.py,sha256=PPpJLowJhgoRijsF2ebmdkFbIriI_yIFlCi1wL6hSP8,42267
-spiderforce4ai/post_extraction_agent.py,sha256=t9KxjuNw16-6kige6ULPLyykNkiGmKhpCi8QjskdaTk,11959
-spiderforce4ai-2.4.5.dist-info/METADATA,sha256=q3VBuGb5wxsi9OPkzEMwFMyg9f_vT2RamWYIgu2JbLc,9012
-spiderforce4ai-2.4.5.dist-info/WHEEL,sha256=In9FTNxeP60KnTkGw7wk6mJPYd_dQSjEZmXdBdMCI-8,91
-spiderforce4ai-2.4.5.dist-info/entry_points.txt,sha256=ibARQxOlDiL1ho12zbDZt4Uq5RKSIk_qk159ZlZ46hc,59
-spiderforce4ai-2.4.5.dist-info/top_level.txt,sha256=Kth7A21Js7DCp0j5XBBi-FE45SCLouZkeNZU__Yr9Yk,15
-spiderforce4ai-2.4.5.dist-info/RECORD,,