Update app.py

app.py

@@ -1,58 +1,32 @@
 import streamlit as st
-from docling.parsers import DOCXParser, PDFParser, PPTXParser
-from docling.utils import to_json, to_markdown
-import sys
-import traceback
+import logging
+from pathlib import Path
+from docling.document_converter import DocumentConverter
+from docling.datamodel.base_models import InputFormat
+from docling.datamodel.pipeline_options import PdfPipelineOptions
+from docling.document_converter import PdfFormatOption
 
-def init_parsers():
-    parsers = {
-        "pdf": PDFParser(),
-        "docx": DOCXParser(),
-        "pptx": PPTXParser()
-    }
-    return parsers
-
-def process_document(file, parser):
+def test_docling():
     try:
-        document = parser.parse(file)
-        return document, None
+        pipeline_options = PdfPipelineOptions()
+        pipeline_options.do_ocr = False
+        pipeline_options.do_table_structure = True
+        
+        converter = DocumentConverter(
+            format_options={
+                InputFormat.PDF: PdfFormatOption(pipeline_options=pipeline_options)
+            }
+        )
+        return True, "Docling setup successful"
     except Exception as e:
-        error_msg = f"Error: {str(e)}\n{traceback.format_exc()}"
-        return None, error_msg
-
-def main():
-    st.set_page_config(page_title="Document Processor")
-    st.title("Docling Document Processor")
-    
-    try:
-        parsers = init_parsers()
-    except ImportError as e:
-        st.error("Error: Required packages not installed. Please run: pip install docling python-docx pdfminer.six python-pptx")
-        return
+        return False, f"Error: {str(e)}"
 
-    uploaded_file = st.file_uploader("Choose a document", type=list(parsers.keys()))
-    
-    if uploaded_file:
-        file_extension = uploaded_file.name.split(".")[-1].lower()
-        parser = parsers.get(file_extension)
-        
-        document, error = process_document(uploaded_file, parser)
-        
-        if error:
-            st.error(f"Failed to parse document:\n{error}")
-            return
-            
-        output_format = st.radio("Select output format:", ("Markdown", "JSON"))
-        
-        try:
-            if output_format == "Markdown":
-                st.subheader("Markdown Output:")
-                st.markdown(to_markdown(document))
-            else:
-                st.subheader("JSON Output:")
-                st.json(to_json(document))
-        except Exception as e:
-            st.error(f"Error converting document: {str(e)}")
+st.title("Docling Test App")
 
-if __name__ == "__main__":
-    main()
+if st.button("Test Docling Setup"):
+    with st.spinner("Testing Docling installation..."):
+        success, message = test_docling()
+        if success:
+            st.success(message)
+        else:
+            st.error(message)