spiderforce4ai 2.5.7__tar.gz → 2.5.9__tar.gz

{spiderforce4ai-2.5.7 → spiderforce4ai-2.5.9}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: spiderforce4ai
-Version: 2.5.7
+Version: 2.5.9
 Summary: Python wrapper for SpiderForce4AI HTML-to-Markdown conversion service with LLM post-processing
 Home-page: https://petertam.pro
 Author: Piotr Tamulewicz

{spiderforce4ai-2.5.7 → spiderforce4ai-2.5.9}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "spiderforce4ai"
-version = "2.5.7"
+version = "2.5.9"
 description = "Python wrapper for SpiderForce4AI HTML-to-Markdown conversion service with LLM post-processing"
 readme = "README.md"
 authors = [

{spiderforce4ai-2.5.7 → spiderforce4ai-2.5.9}/setup.py

@@ -7,7 +7,7 @@ with open("README.md", encoding="utf-8") as f:
 
 setup(
     name="spiderforce4ai",
-    version="2.5.7",
+    version="2.5.9",
     author="Piotr Tamulewicz",
     author_email="pt@petertam.pro",
     description="Python wrapper for SpiderForce4AI HTML-to-Markdown conversion service with LLM post-processing",

{spiderforce4ai-2.5.7 → spiderforce4ai-2.5.9}/spiderforce4ai/post_extraction_agent.py

@@ -167,6 +167,8 @@ class PostExtractionAgent:
                 backup_path = self.config.output_file.with_suffix(f".bak_{int(time.time())}")
                 self.config.output_file.rename(backup_path)
             self.config.output_file.touch()
+            # Initialize empty results dictionary
+            self.results = {}
 
     def _process_single_content(self, url: str, content: str) -> Optional[Dict]:
         """Process a single piece of content through the LLM."""
@@ -258,13 +260,33 @@ class PostExtractionAgent:
         """Save individual or combined results synchronously."""
         try:
             if self.config.combine_output and self.config.output_file:
+                # Update the results dictionary
                 self.results[url] = result
-                with open(self.config.output_file, 'w') as f:
+                
+                # Ensure output directory exists
+                self.config.output_file.parent.mkdir(parents=True, exist_ok=True)
+                
+                # Save combined results atomically
+                temp_file = self.config.output_file.with_suffix('.tmp')
+                with open(temp_file, 'w') as f:
                     json.dump(self.results, f, indent=2)
+                temp_file.replace(self.config.output_file)
+                
+                logger.info(f"Updated combined results file with {url}")
+                
+                # Cleanup backup files after successful save
+                for backup_file in self.config.output_file.parent.glob(f"{self.config.output_file.stem}.bak_*"):
+                    try:
+                        backup_file.unlink()
+                        logger.info(f"Cleaned up backup file: {backup_file}")
+                    except Exception as e:
+                        logger.warning(f"Failed to remove backup file {backup_file}: {e}")
+                        
             elif not self.config.combine_output and self.config.output_file:
                 individual_file = self.config.output_file.parent / f"{url.replace('/', '_')}.json"
                 with open(individual_file, 'w') as f:
                     json.dump(result, f, indent=2)
+                logger.info(f"Saved individual result file for {url}")
         except Exception as e:
             logger.error(f"Error saving results for {url}: {str(e)}")
 
@@ -311,14 +333,19 @@ class PostExtractionAgent:
                 
                 # Save result synchronously
                 try:
-                    self._save_result_sync(url, result)
-                    logger.info(f"Saved results for {url}")
+                    # Save both original and transformed result
+                    if self.config.custom_transform_function:
+                        self._save_result_sync(url, transformed_result)
+                        logger.info(f"Saved transformed results for {url}")
+                    else:
+                        self._save_result_sync(url, result)
+                        logger.info(f"Saved original 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
+                return transformed_result if self.config.custom_transform_function else result
             
             # Wait before retry
             if attempt < self.config.max_retries - 1:

{spiderforce4ai-2.5.7 → spiderforce4ai-2.5.9}/spiderforce4ai.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: spiderforce4ai
-Version: 2.5.7
+Version: 2.5.9
 Summary: Python wrapper for SpiderForce4AI HTML-to-Markdown conversion service with LLM post-processing
 Home-page: https://petertam.pro
 Author: Piotr Tamulewicz