Update app.py

app.py

@@ -10,7 +10,6 @@ import base64
 
 # Constants
 DEFAULT_MODEL = "opengvlab/internvl3-14b:free"
-# No need for placeholder text as we'll use password type input
 DEFAULT_SITE_URL = "https://dynamic-nature-trail-guide.app"
 DEFAULT_SITE_NAME = "Dynamic Nature Trail Guide"
 
@@ -18,15 +17,12 @@ DEFAULT_SITE_NAME = "Dynamic Nature Trail Guide"
 SYSTEM_PROMPT = """
 You are the Dynamic Nature Trail Guide, an expert in identifying and explaining natural elements 
 found on nature trails. For any image sent, please:
-
 1. Identify all visible plants, animals, geological features, and ecosystems
 2. Provide educational information about identified elements
 3. Mention any seasonal characteristics visible in the image
 4. Note any ecological significance or conservation considerations
-5. Offer suggestions for what to observe or learn about next on the trail
-
+5. Offer suggestions for what to observe or learn next on the trail
 Keep explanations informative yet accessible to people of all ages and backgrounds.
-
 IMPORTANT: Structure your responses with clear sections and headings.
 """
 
@@ -36,7 +32,7 @@ def encode_image_to_base64(image_path):
         return base64.b64encode(image_file.read()).decode('utf-8')
 
 def format_response_as_html(text):
-    """Convert the model's text response to formatted HTML"""
+    """Convert the model's text response to formatted HTML with improved section handling"""
     if not text:
         return ""
     
@@ -44,46 +40,84 @@ def format_response_as_html(text):
     if text.startswith("Error analyzing image:"):
         return f'<div style="color: red; padding: 10px; border: 1px solid red; border-radius: 5px;">{text}</div>'
     
-    # Replace newlines with HTML breaks
-    text = text.replace('\n\n', '</p><p>').replace('\n', '<br>')
+    # Parse sections using regex patterns
+    # Look for patterns like "### 1. Identification of Elements" or "## Plants:"
+    sections = re.split(r'(?:#{1,3}\s*\d*\.?\s*|\*\*)([\w\s]+)(?:\*\*|:)', text)
     
-    # Handle headings - look for patterns like "1. Identification:" or "Species Identified:"
-    heading_patterns = [
-        (r'([A-Za-z\s]+):(?=<br>|</p>)', r'<h3>\1</h3>'),  # "Category:" at start of line
-        (r'(\d+\.\s+[A-Za-z\s]+):(?=<br>|</p>)', r'<h3>\1</h3>'),  # "1. Category:" format
-    ]
+    # If splitting didn't work well, fall back to a simpler approach
+    if len(sections) <= 1:
+        # Simple formatting with paragraphs
+        text = text.replace('\n\n', '</p><p>')
+        return f'<div style="padding: 15px; font-family: Arial, sans-serif; line-height: 1.6;"><p>{text}</p></div>'
     
-    for pattern, replacement in heading_patterns:
-        text = re.sub(pattern, replacement, text)
+    # Build HTML from sections
+    html_parts = ['<div style="padding: 15px; font-family: Arial, sans-serif; line-height: 1.6;">']
     
-    # Enhance species names with bold
-    text = re.sub(r'\b([A-Z][a-z]+\s+[a-z]+)\b(?!\<\/)', r'<strong>\1</strong>', text)
+    for i in range(1, len(sections), 2):
+        if i < len(sections) - 1:
+            section_title = sections[i].strip()
+            section_content = sections[i+1].strip()
+            
+            # Apply formatting to section content
+            section_content = format_section_content(section_content)
+            
+            html_parts.append(f'<div class="section">')
+            html_parts.append(f'<h3 style="color: #336699; border-bottom: 1px solid #ccc; padding-bottom: 5px;">{section_title}</h3>')
+            html_parts.append(f'<div class="content">{section_content}</div>')
+            html_parts.append('</div>')
+    
+    # If we couldn't parse sections properly, use the original text
+    if len(html_parts) == 1:
+        text = format_section_content(text)
+        html_parts.append(f'<p>{text}</p>')
+    
+    html_parts.append('</div>')
+    return ''.join(html_parts)
+
+def format_section_content(content):
+    """Format the content of a section with enhanced styling"""
+    # Replace line breaks with paragraph breaks for better readability
+    content = re.sub(r'\n\s*\n', '</p><p>', content)
+    content = re.sub(r'\n(?!<\/p>)', '<br>', content)
+    
+    # Convert bullet points
+    content = re.sub(r'^\s*-\s*(.+?)$', r'<li>\1</li>', content, flags=re.MULTILINE)
+    content = re.sub(r'(<li>.*?</li>)', r'<ul>\1</ul>', content, flags=re.DOTALL)
+    # Fix nested lists (remove empty ul tags)
+    content = re.sub(r'<ul>\s*</ul>', '', content)
+    
+    # Handle subsections like "**Plants:**"
+    content = re.sub(r'\*\*([\w\s]+):\*\*', r'<h4 style="margin-bottom: 5px; color: #336699;">\1</h4>', content)
+    
+    # Enhance species names with bold and italics for scientific names
+    content = re.sub(r'\b([A-Z][a-z]+\s+[a-z]+)\b(?!\<\/)', r'<strong><em>\1</em></strong>', content)
     
     # Add some color to certain keywords
     color_mappings = {
-        'endangered': 'red',
+        'endangered': '#e74c3c',
         'rare': '#FF6600',
-        'native': '#006600',
+        'native': '#27ae60',
         'invasive': '#CC0000',
-        'ecosystem': '#006699',
+        'ecosystem': '#3498db',
         'habitat': '#336699',
+        'conservation': '#16a085',
+        'protected': '#9b59b6',
+        'biodiversity': '#2ecc71'
     }
     
     for keyword, color in color_mappings.items():
-        text = re.sub(r'\b' + keyword + r'\b', f'<span style="color: {color};">{keyword}</span>', text, flags=re.IGNORECASE)
-    
-    # Wrap the entire content in a styled div
-    html = f'''
-    <div style="padding: 15px; font-family: Arial, sans-serif; line-height: 1.6;">
-        <p>{text}</p>
-    </div>
-    '''
+        content = re.sub(r'\b' + keyword + r'\b', f'<span style="color: {color}; font-weight: bold;">{keyword}</span>', content, flags=re.IGNORECASE)
     
-    return html
+    # Ensure the content starts and ends with paragraph tags if needed
+    if not content.startswith('<p>') and not content.startswith('<ul>') and not content.startswith('<h4'):
+        content = '<p>' + content
+    if not content.endswith('</p>') and not content.endswith('</ul>'):
+        content = content + '</p>'
+        
+    return content
 
 def analyze_image(api_key, image, prompt="What can you identify in this nature trail image? Provide detailed educational information.", site_url=DEFAULT_SITE_URL, site_name=DEFAULT_SITE_NAME, model=DEFAULT_MODEL):
     """Analyze the uploaded image using the InternVL3 model via OpenRouter"""
-    # Remove the placeholder text check
     if not api_key:
         return "<div style='color: red; padding: 10px; border: 1px solid red; border-radius: 5px;'>Please provide an OpenRouter API key.</div>"
     
@@ -276,5 +310,4 @@ def create_interface():
 
 # Create and launch the app
 if __name__ == "__main__":
-    app = create_interface()
-    app.launch()
+    app = create_interface()