switch from docling to markitdown

app.py

@@ -5,12 +5,12 @@ from typing import List
 from PIL import Image
 from loadimg import load_img
 import io
-from docling.document_converter import DocumentConverter
+import markitdown
 
 converter = Converter()
-docling_converter = DocumentConverter()
 
-def convert_file_to_pdf(filename:str) -> str:
+
+def convert_file_to_pdf(filename: str) -> str:
     """
     Converts a markdown file to PDF format.
 
@@ -21,11 +21,12 @@ def convert_file_to_pdf(filename:str) -> str:
     Returns:
         str: The file path of the generated PDF file.
     """
-    output_path = filename.name.rsplit('.', 1)[0] + '.pdf'
+    output_path = filename.name.rsplit(".", 1)[0] + ".pdf"
     converter.convert(filename.name, output_path)
     return output_path
 
-def convert_file_to_img(image_file:str=None,txt:str="") -> List[Image.Image] :
+
+def convert_file_to_img(image_file: str = None, txt: str = "") -> List[Image.Image]:
     """
     Convert an image file to PDF format.
 
@@ -39,10 +40,10 @@ def convert_file_to_img(image_file:str=None,txt:str="") -> List[Image.Image] :
              the same as the input filename but with a .pdf extension.
     """
     img_list = []
-    if txt != "": 
-        img_list.append(load_img(txt,output_type="pil"))
-    if image_file is not None: 
-        output_path = image_file.name.rsplit('.', 1)[0] + '.pdf'
+    if txt != "":
+        img_list.append(load_img(txt, output_type="pil"))
+    if image_file is not None:
+        output_path = image_file.name.rsplit(".", 1)[0] + ".pdf"
         converter.convert(image_file.name, output_path)
         doc = fitz.open(output_path)
         for page in doc:
@@ -50,19 +51,18 @@ def convert_file_to_img(image_file:str=None,txt:str="") -> List[Image.Image] :
             img_list.append(load_img(Image.open(io.BytesIO(page_bytes))).convert("RGB"))
     doc.close()
     return img_list
-    
-    
-def convert_file_to_markdown(filename:str) -> str:
+
+
+def convert_file_to_markdown(filename: str) -> str:
     """
-    Converts a file to markdown format.
+    Converts a file to markdown format using markitdown.
     Args:
         filename: str
             The path to the file to be converted.
     Returns:
         str: The markdown representation of the file.
     """
-    result = docling_converter.convert(filename)
-    return result.document.export_to_markdown()
+    return markitdown.convert(filename.name)
 
 
 # Create individual interfaces
@@ -71,15 +71,15 @@ file_to_pdf = gr.Interface(
     inputs=gr.File(label="Upload README/Markdown file"),
     outputs=gr.File(label="Converted PDF"),
     title="File to PDF Converter",
-    description="Convert your files to PDF format"
+    description="Convert your files to PDF format",
 )
 
 file_to_image = gr.Interface(
     fn=convert_file_to_img,
-    inputs=[gr.File(label="Upload Image"),gr.Textbox(label="base64, url")],
+    inputs=[gr.File(label="Upload Image"), gr.Textbox(label="base64, url")],
     outputs=gr.Gallery(label="Converted Images"),
     title="File to Images Converter",
-    description="Convert your images to an image format"
+    description="Convert your images to an image format",
 )
 
 file_to_markdown = gr.Interface(
@@ -87,7 +87,7 @@ file_to_markdown = gr.Interface(
     inputs=gr.File(label="Upload File"),
     outputs=gr.Textbox(label="Converted Markdown"),
     title="File to Markdown Converter",
-    description="Convert your files to markdown format"
+    description="Convert your files to markdown format",
 )
 
 # Create tabbed interface
@@ -97,4 +97,4 @@ demo = gr.TabbedInterface(
 )
 
 if __name__ == "__main__":
-    demo.launch(server_name="0.0.0.0", server_port=7860, debug=True, mcp_server=True)
+    demo.launch(server_name="0.0.0.0", server_port=7860, debug=True, mcp_server=True)

requirements.txt

@@ -8,4 +8,5 @@ gradio
 python-pptx
 pdfitdown
 loadimg
-docling
+docling
+markitdown