Update app.py

app.py

@@ -1,146 +1,56 @@
-# app.py
-
 import gradio as gr
-from markitdown import MarkItDown
-from pathlib import Path
-import tempfile
+import fitz  # PyMuPDF
+import docx
+from markdownify import markdownify as md
 import os
-import traceback
-
-# Suggested limit — adjust as needed (Hugging Face Spaces free tier often allows 500 MB+)
-MAX_FILE_SIZE_STR = "100MB"   # or "500MB", "2GB", etc.
-
-def convert_to_markdown(file_obj):
-    if file_obj is None:
-        return "Please upload a document.", ""
-
-    try:
-        original_name = Path(file_obj.name).stem
-        ext = Path(file_obj.name).suffix.lower()
-
-        allowed_extensions = [
-            '.pdf', '.doc', '.docx', '.ppt', '.pptx',
-            '.xls', '.xlsx', '.odt', '.rtf', '.txt', '.md'
-        ]
-
-        if ext not in allowed_extensions:
-            return (
-                f"Unsupported file format: {ext}\n\n"
-                f"Supported: {', '.join(allowed_extensions)}",
-                ""
-            )
-
-        # Write uploaded content to temporary file
-        with tempfile.NamedTemporaryFile(delete=False, suffix=ext) as tmp:
-            # Important: file_obj is a NamedTemporaryFile already in recent Gradio → read() works
-            tmp.write(file_obj.read())
-            tmp_path = tmp.name
-
-        try:
-            md = MarkItDown()
-            result = md.convert(tmp_path)
-            markdown_text = result.text_content
-
-            suggested_filename = f"{original_name}.md"
+import tempfile
 
-            return (
-                f"**Conversion successful!** ✓  \nFile: {original_name}{ext}",
-                markdown_text,
-                suggested_filename
-            )
+def extract_text_from_pdf(file_path):
+    text = ""
+    doc = fitz.open(file_path)
+    for page in doc:
+        text += page.get_text()
+    return text
 
-        finally:
-            try:
-                os.unlink(tmp_path)
-            except:
-                pass
+def extract_text_from_docx(file_path):
+    doc = docx.Document(file_path)
+    return "\n".join([p.text for p in doc.paragraphs])
 
-    except Exception as e:
-        error_detail = traceback.format_exc()
-        return f"**Error during conversion:**\n{str(e)}\n\n{error_detail[:1000]}…", "", ""
+def convert_to_markdown(uploaded_file):
+    if uploaded_file is None:
+        return None
 
+    file_path = uploaded_file.name
+    ext = os.path.splitext(file_path)[1].lower()
 
-css = """
-.upload-box {border: 1px solid #ccc; border-radius: 8px; padding: 16px; background: #fafafa;}
-.success {color: #2e7d32;}
-.error   {color: #c62828;}
-"""
+    if ext == ".pdf":
+        text = extract_text_from_pdf(file_path)
+    elif ext == ".docx":
+        text = extract_text_from_docx(file_path)
+    else:
+        return None
 
-with gr.Blocks(title="Document → Markdown Converter") as demo:   # ← removed css= here
+    markdown_text = md(text)
 
-    gr.Markdown("""
-    # Document → Markdown Converter
+    temp_md = tempfile.NamedTemporaryFile(delete=False, suffix=".md")
+    with open(temp_md.name, "w", encoding="utf-8") as f:
+        f.write(markdown_text)
 
-    Upload PDF, Word, PowerPoint, Excel, etc. → clean Markdown
+    return temp_md.name
 
-    Powered by **markitdown** • Best for text-heavy documents
-    """)
+with gr.Blocks() as demo:
+    gr.Markdown("# 📄➡️📝 Document to Markdown Converter")
+    gr.Markdown("Upload a **PDF or Word (.docx)** file and get a **Markdown (.md)** file.")
 
-    with gr.Row():
-        with gr.Column(scale=4):
-            file_input = gr.File(
-                label="Upload document",
-                file_count="single",
-                file_types=[
-                    ".pdf", ".doc", ".docx", ".ppt", ".pptx",
-                    ".xls", ".xlsx", ".odt", ".rtf", ".txt", ".md"
-                ],
-                elem_classes="upload-box",
-                # NO max_size here — it was never valid
-            )
+    file_input = gr.File(label="Upload PDF or DOCX", file_types=[".pdf", ".docx"])
+    output_file = gr.File(label="Download Markdown File")
 
-        with gr.Column(scale=1, min_width=180):
-            convert_btn = gr.Button("Convert to Markdown", variant="primary")
+    convert_btn = gr.Button("Convert to Markdown")
 
-    status_output = gr.Textbox(
-        label="Status",
-        lines=3,
-        interactive=False,
-        show_copy_button=False
-    )
-
-    markdown_output = gr.Markdown(
-        label="Converted Markdown",
-        height=500
-    )
-
-    with gr.Row():
-        download_file = gr.File(
-            label="Download markdown file",
-            file_types=[".md"],
-            interactive=False
-        )
-        download_name = gr.Textbox(
-            label="Suggested filename",
-            value="document.md",
-            interactive=True,
-            max_lines=1
-        )
-
-    # Main action
     convert_btn.click(
         fn=convert_to_markdown,
         inputs=file_input,
-        outputs=[status_output, markdown_output, download_name]
-    ).then(
-        fn=lambda md_text, fname: gr.File(value=md_text, filename=fname) if md_text else None,
-        inputs=[markdown_output, download_name],
-        outputs=download_file
+        outputs=output_file
     )
 
-    gr.Markdown(f"""
-    ### Notes
-    - Maximum file size: **{MAX_FILE_SIZE_STR}** (set via launch parameter)
-    - Best with text-oriented documents
-    - Tables / images / layouts may be simplified
-    - Very large files may timeout or fail
-    """)
-
-
-if __name__ == "__main__":
-    demo.launch(
-        max_file_size=MAX_FILE_SIZE_STR,   # ← This is where the limit goes!
-        # Optional: in_spaces you often want these too
-        # server_name="0.0.0.0",
-        # server_port=7860,
-    )
+demo.launch()