Create app.py

app.py

@@ -0,0 +1,48 @@
+import gradio as gr
+import subprocess
+import os
+import uuid
+import shutil
+
+def convert_pdf_to_html(pdf_file):
+    job_id = str(uuid.uuid4())
+    work_dir = f"/tmp/{job_id}"
+    os.makedirs(work_dir, exist_ok=True)
+
+    input_path = os.path.join(work_dir, "input.pdf")
+    output_path = os.path.join(work_dir, "output.html")
+
+    with open(input_path, "wb") as f:
+        f.write(pdf_file.read())
+
+    try:
+        # Run pdf2htmlEX
+        result = subprocess.run(
+            ["pdf2htmlEX", "--dest-dir", work_dir, "--embed", "cfijo", input_path],
+            capture_output=True,
+            text=True,
+        )
+        if result.returncode != 0:
+            return f"Error during conversion:\n{result.stderr}"
+
+        # Find the output HTML file (usually named input.html)
+        converted_file = os.path.join(work_dir, "input.html")
+        if not os.path.exists(converted_file):
+            return "Conversion failed: HTML file not found."
+
+        return converted_file
+
+    finally:
+        # Optional: remove temp files after Gradio session ends
+        shutil.rmtree(work_dir, ignore_errors=True)
+
+demo = gr.Interface(
+    fn=convert_pdf_to_html,
+    inputs=gr.File(label="Upload PDF", file_types=[".pdf"]),
+    outputs=gr.File(label="Download HTML"),
+    title="PDF to HTML Converter (pdf2htmlEX)",
+    description="Upload a PDF and download an HTML version using pdf2htmlEX."
+)
+
+if __name__ == "__main__":
+    demo.launch()