spiderforce4ai 2.4.3__tar.gz → 2.4.5__tar.gz

{spiderforce4ai-2.4.3 → spiderforce4ai-2.4.5}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: spiderforce4ai
-Version: 2.4.3
+Version: 2.4.5
 Summary: Python wrapper for SpiderForce4AI HTML-to-Markdown conversion service with LLM post-processing
 Home-page: https://petertam.pro
 Author: Piotr Tamulewicz

{spiderforce4ai-2.4.3 → spiderforce4ai-2.4.5}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "spiderforce4ai"
-version = "2.4.3"
+version = "2.4.5"
 description = "Python wrapper for SpiderForce4AI HTML-to-Markdown conversion service with LLM post-processing"
 readme = "README.md"
 authors = [

{spiderforce4ai-2.4.3 → spiderforce4ai-2.4.5}/setup.py

@@ -7,7 +7,7 @@ with open("README.md", encoding="utf-8") as f:
 
 setup(
     name="spiderforce4ai",
-    version="2.4.3",
+    version="2.4.5",
     author="Piotr Tamulewicz",
     author_email="pt@petertam.pro",
     description="Python wrapper for SpiderForce4AI HTML-to-Markdown conversion service with LLM post-processing",

{spiderforce4ai-2.4.3 → spiderforce4ai-2.4.5}/spiderforce4ai/__init__.py

@@ -586,7 +586,7 @@ class SpiderForce4AI:
                 for result in results:
                     if result.status == "success":
                         try:
-                            result.extraction_result = await agent.process_content(result.url, result.markdown)
+                            result.extraction_result = agent.process_content(result.url, result.markdown)
                             progress.update(llm_task, advance=1)
                         except Exception as e:
                             console.print(f"[red]Error in post-extraction processing for {result.url}: {str(e)}[/red]")

{spiderforce4ai-2.4.3 → spiderforce4ai-2.4.5}/spiderforce4ai/post_extraction_agent.py

@@ -164,12 +164,9 @@ class PostExtractionAgent:
                 self.config.output_file.rename(backup_path)
             self.config.output_file.touch()
 
-    async def _process_single_content(self, url: str, content: str) -> Optional[Dict]:
+    def _process_single_content(self, url: str, content: str) -> Optional[Dict]:
         """Process a single piece of content through the LLM."""
         try:
-            # Apply rate limiting
-            await self.rate_limiter.acquire()
-            
             # Replace placeholder in messages with actual content
             messages = [
                 {**msg, 'content': msg['content'].replace('{here_markdown_content}', content)}
@@ -183,7 +180,8 @@ class PostExtractionAgent:
             
             for attempt in range(max_retries):
                 try:
-                    response = await completion(
+                    # Call completion synchronously
+                    response = completion(
                         model=self.config.model,
                         messages=messages,
                         max_tokens=self.config.max_tokens,
@@ -200,11 +198,11 @@ class PostExtractionAgent:
                 except json.JSONDecodeError as e:
                     last_error = f"Invalid JSON response from LLM: {e}"
                     if attempt < max_retries - 1:
-                        await asyncio.sleep(retry_delay * (attempt + 1))
+                        time.sleep(retry_delay * (attempt + 1))
                 except Exception as e:
                     last_error = str(e)
                     if attempt < max_retries - 1:
-                        await asyncio.sleep(retry_delay * (attempt + 1))
+                        time.sleep(retry_delay * (attempt + 1))
             
             # If we get here, all retries failed
             raise Exception(last_error)
@@ -214,6 +212,20 @@ class PostExtractionAgent:
             self.buffer.add_failed_request(url, content, str(e))
             return None
 
+    def _save_result_sync(self, url: str, result: Dict) -> None:
+        """Save individual or combined results synchronously."""
+        try:
+            if self.config.combine_output and self.config.output_file:
+                self.results[url] = result
+                with open(self.config.output_file, 'w') as f:
+                    json.dump(self.results, f, indent=2)
+            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)
+        except Exception as e:
+            logger.error(f"Error saving results for {url}: {str(e)}")
+
     async def _save_result(self, url: str, result: Dict) -> None:
         """Save individual or combined results."""
         try:
@@ -228,10 +240,10 @@ class PostExtractionAgent:
         except Exception as e:
             logger.error(f"Error saving results for {url}: {str(e)}")
 
-    async def process_content(self, url: str, content: str) -> Optional[Dict]:
+    def process_content(self, url: str, content: str) -> Optional[Dict]:
         """Process content with retry mechanism."""
         for attempt in range(self.config.max_retries):
-            result = await self._process_single_content(url, content)
+            result = self._process_single_content(url, content)
             if result:
                 # Apply custom transformation if provided
                 if self.config.custom_transform_function:
@@ -240,12 +252,13 @@ class PostExtractionAgent:
                     except Exception as e:
                         logger.error(f"Error in custom transform for {url}: {str(e)}")
                 
-                await self._save_result(url, result)
+                # Save result synchronously
+                self._save_result_sync(url, result)
                 return result
             
             # Wait before retry
             if attempt < self.config.max_retries - 1:
-                await asyncio.sleep(self.config.retry_delay)
+                time.sleep(self.config.retry_delay)
         
         return None
 

{spiderforce4ai-2.4.3 → spiderforce4ai-2.4.5}/spiderforce4ai.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: spiderforce4ai
-Version: 2.4.3
+Version: 2.4.5
 Summary: Python wrapper for SpiderForce4AI HTML-to-Markdown conversion service with LLM post-processing
 Home-page: https://petertam.pro
 Author: Piotr Tamulewicz