Update app.py

app.py

@@ -24,57 +24,45 @@ def convert_pdf_to_images(pdf_file):
     return images
 
 def format_response(text):
+    """Format the analysis response with clean styling"""
     formatted_text = """
     <div style="
         background-color: white;
-        padding: 30px;
-        border-radius: 10px;
+        padding: 20px;
+        border-radius: 5px;
         font-family: Arial, sans-serif;
-        box-shadow: 0 2px 8px rgba(0,0,0,0.1);
-        margin: 20px 0;
+        box-shadow: 0 2px 4px rgba(0,0,0,0.1);
     ">
     """
     
+    # Split into pages
     pages = text.split("Page")
     
-    for page_num, page_content in enumerate(pages[1:], 1):
-        formatted_text += f'''
-            <div style="margin-bottom: 40px;">
-                <h3 style="color: #2c3e50; 
-                          margin-bottom: 20px;
-                          padding-bottom: 10px;
-                          border-bottom: 2px solid #eee;">
-                    Page {page_num}
-                </h3>
-        '''
+    for page_num, page_content in enumerate(pages[1:], 1):  # Skip first empty split
+        # Add page header
+        formatted_text += f'<div style="margin-bottom: 30px;">'
+        formatted_text += f'<h3 style="color: #2c3e50; margin-bottom: 15px;">Page {page_num}</h3>'
         
+        # Process each line
         lines = page_content.split('\n')
         for line in lines:
+            # Skip empty lines and lines with asterisks
             if line.strip() and not line.strip().startswith('*') and not line.strip().startswith('Here'):
+                # Remove asterisks and dashes
                 line = line.replace('**', '').replace('- ', '')
                 
                 if ':' in line:
                     label, value = line.split(':', 1)
-                    formatted_text += f'''
-                        <div style="margin-bottom: 15px; 
-                                  display: flex;
-                                  align-items: baseline;
-                                  gap: 20px;">
-                            <span style="font-weight: 600; 
-                                     color: #2c3e50; 
-                                     min-width: 200px;">
-                                {label.strip()}
-                            </span>
-                            <span style="color: #34495e; flex: 1;">
-                                {value.strip()}
-                            </span>
-                        </div>
-                    '''
+                    formatted_text += f'<div style="margin-bottom: 10px; display: flex;">'
+                    formatted_text += f'<span style="font-weight: bold; color: #2c3e50; min-width: 200px;">{label.strip()}</span>'
+                    formatted_text += f'<span style="color: #34495e; flex: 1;">{value.strip()}</span>'
+                    formatted_text += '</div>'
         
         formatted_text += '</div>'
         
+        # Add separator between pages except for the last page
         if page_num < len(pages) - 1:
-            formatted_text += '<hr style="border: none; border-top: 1px solid #eee; margin: 30px 0;">'
+            formatted_text += '<hr style="border: 1px solid #eee; margin: 20px 0;">'
     
     formatted_text += '</div>'
     return formatted_text
@@ -89,7 +77,7 @@ def analyze_image(image):
         base64_image = base64.b64encode(img_byte_arr).decode("utf-8")
         
         response = client.chat.completions.create(
-            model="gpt-4-vision-preview",
+            model="gpt-4o-mini",
             messages=[
                 {
                     "role": "user",
@@ -125,16 +113,11 @@ def analyze_image(image):
 def main():
     st.set_page_config(page_title="Document Analysis App", layout="wide")
     
-    # Custom CSS
+    # Custom CSS to set light background and improve button styling
     st.markdown("""
         <style>
         .stApp {
-            background-color: #ffffff !important;
-        }
-        .main {
-            padding: 20px;
-            max-width: 1200px;
-            margin: 0 auto;
+            background-color: white;
         }
         .stButton>button {
             width: 100%;
@@ -143,68 +126,48 @@ def main():
             border: none;
             padding: 10px 20px;
             border-radius: 5px;
-            margin: 20px 0;
+            margin-top: 20px;
+            margin-bottom: 20px;
         }
         .stButton>button:hover {
             background-color: #34495e;
         }
         .uploadedFile {
-            margin: 20px 0;
-            padding: 20px;
-            background-color: #f8f9fa;
-            border-radius: 5px;
-        }
-        [data-testid="stHeader"] {
-            background-color: transparent;
-        }
-        .stDeployButton {
-            display: none !important;
-        }
-        .main .block-container {
-            padding-top: 2rem;
-            padding-bottom: 2rem;
+            margin-bottom: 20px;
         }
         </style>
     """, unsafe_allow_html=True)
     
-    # Main container
-    with st.container():
-        col1, col2, col3 = st.columns([1,3,1])
+    col1, col2, col3 = st.columns([1,2,1])
+    with col2:
+        st.title("Document Analysis App")
         
-        with col2:
-            st.title("Document Analysis App")
-            
-            upload_container = st.container()
-            with upload_container:
-                uploaded_file = st.file_uploader(
-                    "Upload document (PDF/Image)", 
-                    type=['pdf', 'png', 'jpg', 'jpeg']
-                )
-            
-            if uploaded_file is not None:
-                if uploaded_file.type == "application/pdf":
-                    # Handle PDF
-                    with st.spinner("Processing PDF..."):
-                        images = convert_pdf_to_images(uploaded_file)
-                        
-                        if st.button("Extract Information"):
-                            with st.spinner("Analyzing document..."):
-                                all_results = []
-                                for i, image in enumerate(images, 1):
-                                    result = analyze_image(image)
-                                    all_results.append(f"Page {i} Information:\n{result}")
-                                
-                                combined_results = "\n\n".join(all_results)
-                                st.markdown(format_response(combined_results), unsafe_allow_html=True)
-                                
-                else:
-                    # Handle single image
-                    image = Image.open(uploaded_file)
+        uploaded_file = st.file_uploader("Upload document (PDF/Image)", type=['pdf', 'png', 'jpg', 'jpeg'])
+        
+        if uploaded_file is not None:
+            if uploaded_file.type == "application/pdf":
+                # Handle PDF
+                with st.spinner("Processing PDF..."):
+                    images = convert_pdf_to_images(uploaded_file)
                     
                     if st.button("Extract Information"):
                         with st.spinner("Analyzing document..."):
-                            result = analyze_image(image)
-                            st.markdown(format_response(f"Page 1\n{result}"), unsafe_allow_html=True)
+                            all_results = []
+                            for i, image in enumerate(images, 1):
+                                result = analyze_image(image)
+                                all_results.append(f"Page {i} Information:\n{result}")
+                            
+                            combined_results = "\n\n".join(all_results)
+                            st.markdown(format_response(combined_results), unsafe_allow_html=True)
+                            
+            else:
+                # Handle single image
+                image = Image.open(uploaded_file)
+                
+                if st.button("Extract Information"):
+                    with st.spinner("Analyzing document..."):
+                        result = analyze_image(image)
+                        st.markdown(format_response(result), unsafe_allow_html=True)
 
 if __name__ == "__main__":
     main()