Update modules/preprocessing.py

modules/preprocessing.py

@@ -103,16 +103,22 @@ class PDFProcessor:
         """
         pdf_file = Path(file_path)
         doc = DocumentFile.from_pdf(pdf_file)
-        if len(doc.pages) > self.max_pages:
-            doc.pages = doc.pages[:self.max_pages]
-
-        result = self.ocr_model(doc)
+        
+        # Asegúrate de que `doc` sea un objeto compatible con pages
+        if isinstance(doc, list):
+            pages = doc[:self.max_pages] if len(doc) > self.max_pages else doc
+        elif hasattr(doc, "pages"):
+            pages = doc.pages[:self.max_pages] if len(doc.pages) > self.max_pages else doc.pages
+        else:
+            raise ValueError("Formato inesperado para el documento PDF.")
+        
         raw_text = "\n".join(
-            [block.text for page in result.pages for block in page.blocks]
+            [block.text for page in pages for block in page.blocks]
         )
         return Preprocessor().preprocess_text(raw_text)
 
 
+
 class FileHandler:
     """Clase para manejar archivos temporales y limpieza."""