Update app.py

app.py

@@ -15,41 +15,29 @@ def extract_text_markdown(doc):
         blocks = page.get_text("dict")["blocks"]
         elements = []
 
-        # Extraer imágenes y guardar para asignar link
-        image_list = page.get_images(full=True)
-        xref_to_placeholder = {}
-
-        for img in image_list:
-            xref = img[0]
-            pix = fitz.Pixmap(doc, xref)
-            img_path = os.path.join(image_dir, f"imagen_{image_counter}.png")
-
-            if pix.n > 4:
-                pix = fitz.Pixmap(fitz.csRGB, pix)
-            pix.save(img_path)
-            pix = None
-
-            xref_to_placeholder[xref] = f"![imagen]()"
-            image_counter += 1
-
         for b in blocks:
             y = b["bbox"][1]
             if b["type"] == 0:  # Texto
-                paragraph = ""
                 for line in b["lines"]:
-                    line_text = " ".join([span["text"].strip() for span in line["spans"]])
-                    paragraph += line_text + " "
-                paragraph = paragraph.strip()
-                if paragraph:
-                    elements.append((y, paragraph))
+                    line_y = line["bbox"][1]
+                    line_text = " ".join([span["text"] for span in line["spans"]]).strip()
+                    if line_text:
+                        elements.append((line_y, line_text))
             elif b["type"] == 1:  # Imagen
-                xref = b.get("image")
                 elements.append((y, "![imagen]()"))
 
+        # Ordenar elementos por posición vertical
         elements.sort(key=lambda x: x[0])
 
-        for _, content in elements:
-            markdown_output += content + "\n\n"
+        # Reconstruir markdown respetando el espaciado visual
+        previous_y = None
+        for y, content in elements:
+            if previous_y is not None and abs(y - previous_y) > 10:  # si hay espacio entre líneas, añadir salto
+                markdown_output += "\n"
+            markdown_output += content + "\n"
+            previous_y = y
+
+        markdown_output += "\n---\n\n"  # separador entre páginas
 
     return markdown_output.strip()