Update app.py

app.py

@@ -56,10 +56,76 @@ async def generate_audio(text, voice, filename):
     await communicate.save(filename)
     return filename
 
+# 🔗 Detecting links like a digital bloodhound
+def detect_and_convert_links(text):
+    """🕸️ Finds URLs in your text and turns them into actual clickable links"""
+    # Pattern to find URLs (http/https/ftp/www)
+    url_pattern = re.compile(
+        r'(https?://|www\.)[^\s\[\]()<>{}]+(\.[^\s\[\]()<>{}]+)+(/[^\s\[\]()<>{}]*)?',
+        re.IGNORECASE
+    )
+    
+    # Pattern to find markdown links [text](url)
+    md_link_pattern = re.compile(r'\[(.*?)\]\((https?://[^\s\[\]()<>{}]+)\)')
+    
+    # First convert markdown links
+    text = md_link_pattern.sub(r'<a href="\2">\1</a>', text)
+    
+    # Then find and convert plain URLs not already in tags
+    start_idx = 0
+    result = []
+    
+    while start_idx < len(text):
+        # Find the next URL
+        match = url_pattern.search(text, start_idx)
+        if not match:
+            # No more URLs, add the remaining text
+            result.append(text[start_idx:])
+            break
+        
+        # Check if the URL is already inside an <a> tag
+        # This is a simplified check. A more robust approach might use an HTML parser
+        prev_text = text[start_idx:match.start()]
+        tag_balance = prev_text.count('<a') - prev_text.count('</a')
+        
+        if tag_balance > 0:
+            # URL is likely inside a tag, skip this match
+            result.append(text[start_idx:match.end()])
+        else:
+            # Add text before the URL
+            result.append(text[start_idx:match.start()])
+            
+            # Get the URL
+            url = match.group(0)
+            
+            # Add proper URL prefix if needed
+            if url.startswith('www.'):
+                url_with_prefix = 'http://' + url
+            else:
+                url_with_prefix = url
+            
+            # Add the URL as a link
+            result.append(f'<a href="{url_with_prefix}">{url}</a>')
+        
+        start_idx = match.end()
+    
+    return ''.join(result)
+
 # 🎭 Making emojis wear the right font costume
 def apply_emoji_font(text, emoji_font):
     """🦄 Because emojis deserve their own font fashion show"""
-    # First, handle any bold formatting to avoid tag nesting issues
+    # First handle links - temporarily replace them with placeholders
+    link_pattern = re.compile(r'<a\s+href="([^"]+)">(.*?)</a>')
+    links = []
+    
+    def save_link(match):
+        link_idx = len(links)
+        links.append((match.group(1), match.group(2)))
+        return f"###LINK_{link_idx}###"
+    
+    text = link_pattern.sub(save_link, text)
+    
+    # Now handle bold formatting
     text = re.sub(r'<b>(.*?)</b>', lambda m: f'###BOLD_START###{m.group(1)}###BOLD_END###', text)
     
     # Apply emoji font replacement
@@ -102,6 +168,23 @@ def apply_emoji_font(text, emoji_font):
     combined_text = combined_text.replace('###BOLD_START###', '</font><b><font face="DejaVuSans">')
     combined_text = combined_text.replace('###BOLD_END###', '</font></b><font face="DejaVuSans">')
     
+    # Restore links
+    for i, (url, label) in enumerate(links):
+        placeholder = f"###LINK_{i}###"
+        if placeholder in combined_text:
+            # If the link is within a font tag, we need to close and reopen it
+            parts = combined_text.split(placeholder)
+            if len(parts) == 2:
+                before, after = parts
+                # Check if we're inside a font tag
+                if before.rfind('<font') > before.rfind('</font>'):
+                    # Close font tag before link, reopen after
+                    link_html = f'</font><a href="{url}">{label}</a><font face="DejaVuSans">'
+                    combined_text = before + link_html + after
+                else:
+                    # Simple replacement
+                    combined_text = before + f'<a href="{url}">{label}</a>' + after
+    
     return combined_text
 
 # 📝 Converting markdown to PDF content, because PDFs never go out of style
@@ -115,8 +198,11 @@ def markdown_to_pdf_content(markdown_text, render_with_bold, auto_bold_numbers):
         line = line.strip()
         if not line or line.startswith('# '):
             continue
+        
+        # Process links before any other formatting
+        line = detect_and_convert_links(line)
             
-        # Handle bold formatting before any other processing
+        # Handle bold formatting
         if render_with_bold:
             line = re.sub(r'\*\*(.*?)\*\*', r'<b>\1</b>', line)
             
@@ -147,19 +233,26 @@ def create_pdf(markdown_text, base_font_size, render_with_bold, auto_bold_number
     styles = getSampleStyleSheet()
     spacer_height = 10
     pdf_content, total_lines = markdown_to_pdf_content(markdown_text, render_with_bold, auto_bold_numbers)
+    
+    # Define styles for different text types
     item_style = ParagraphStyle(
         'ItemStyle', parent=styles['Normal'], fontName="DejaVuSans",
-        fontSize=base_font_size, leading=base_font_size * 1.15, spaceAfter=1
+        fontSize=base_font_size, leading=base_font_size * 1.15, spaceAfter=1,
+        linkUnderline=True  # Enable underline for links
     )
     numbered_bold_style = ParagraphStyle(
         'NumberedBoldStyle', parent=styles['Normal'], fontName="NotoEmoji-Bold",
         fontSize=base_font_size + 1 if enlarge_numbered else base_font_size,
-        leading=(base_font_size + 1) * 1.15 if enlarge_numbered else base_font_size * 1.15, spaceAfter=1
+        leading=(base_font_size + 1) * 1.15 if enlarge_numbered else base_font_size * 1.15, spaceAfter=1,
+        linkUnderline=True  # Enable underline for links
     )
     section_style = ParagraphStyle(
         'SectionStyle', parent=styles['Heading2'], fontName="DejaVuSans",
-        textColor=colors.darkblue, fontSize=base_font_size * 1.1, leading=base_font_size * 1.32, spaceAfter=2
+        textColor=colors.darkblue, fontSize=base_font_size * 1.1, leading=base_font_size * 1.32, spaceAfter=2,
+        linkUnderline=True  # Enable underline for links
     )
+    
+    # Register fonts
     try:
         available_font_files = glob.glob("*.ttf")
         if not available_font_files:
@@ -176,17 +269,22 @@ def create_pdf(markdown_text, base_font_size, render_with_bold, auto_bold_number
     except Exception as e:
         st.error(f"Font registration error: {e}")
         return
+        
+    # Distribute content across columns
     columns = [[] for _ in range(num_columns)]
     lines_per_column = total_lines / num_columns if num_columns > 0 else total_lines
     current_line_count = 0
     current_column = 0
     number_pattern = re.compile(r'^\d+\.\s')
+    
     for item in pdf_content:
         if current_line_count >= lines_per_column and current_column < num_columns - 1:
             current_column += 1
             current_line_count = 0
         columns[current_column].append(item)
         current_line_count += 1
+        
+    # Format columns into Paragraph objects
     column_cells = [[] for _ in range(num_columns)]
     for col_idx, column in enumerate(columns):
         for item in column:
@@ -198,9 +296,13 @@ def create_pdf(markdown_text, base_font_size, render_with_bold, auto_bold_number
                     column_cells[col_idx].append(Paragraph(apply_emoji_font(content, "NotoEmoji-Bold"), section_style))
             else:
                 column_cells[col_idx].append(Paragraph(apply_emoji_font(item, "DejaVuSans"), item_style))
+                
+    # Ensure columns have the same number of cells
     max_cells = max(len(cells) for cells in column_cells) if column_cells else 0
     for cells in column_cells:
         cells.extend([Paragraph("", item_style)] * (max_cells - len(cells)))
+        
+    # Create the table layout
     col_width = (page_width - 72) / num_columns if num_columns > 0 else page_width - 72
     table_data = list(zip(*column_cells)) if column_cells else [[]]
     table = Table(table_data, colWidths=[col_width] * num_columns, hAlign='CENTER')
@@ -215,6 +317,8 @@ def create_pdf(markdown_text, base_font_size, render_with_bold, auto_bold_number
         ('TOPPADDING', (0, 0), (-1, -1), 1),
         ('BOTTOMPADDING', (0, 0), (-1, -1), 1),
     ]))
+    
+    # Build the PDF
     story = [Spacer(1, spacer_height), table]
     doc.build(story)
     buffer.seek(0)