Update app.py

app.py

@@ -1,6 +1,8 @@
 import io
 import re
 import streamlit as st
+import glob
+import os
 from PIL import Image
 import fitz
 from reportlab.lib.pagesizes import A4
@@ -13,22 +15,22 @@ from reportlab.pdfbase.ttfonts import TTFont
 st.set_page_config(layout="wide", initial_sidebar_state="collapsed")
 
 def create_pdf_tab(default_markdown):
-    # Font setup
-    available_fonts = {
-        "NotoEmoji Variable": "NotoEmoji-VariableFont_wght.ttf",
-        "NotoEmoji Bold": "NotoEmoji-Bold.ttf",
-        "NotoEmoji Light": "NotoEmoji-Light.ttf",
-        "NotoEmoji Medium": "NotoEmoji-Medium.ttf",
-        "NotoEmoji Regular": "NotoEmoji-Regular.ttf",
-        "NotoEmoji SemiBold": "NotoEmoji-SemiBold.ttf"
-    }
+    # Dynamically load all .ttf fonts from the current directory
+    # Fonts are sourced from: https://fonts.google.com/download/next-steps
+    # We use glob to find all .ttf files, making the font list dynamic instead of hardcoded.
+    font_files = glob.glob("*.ttf")
+    if not font_files:
+        st.error("No .ttf font files found in the current directory. Please add some, e.g., NotoColorEmoji-Regular.ttf.")
+        return
+    available_fonts = {os.path.splitext(os.path.basename(f))[0]: f for f in font_files}
 
     # Sidebar configuration
     with st.sidebar:
-        selected_font_name = st.selectbox("Select NotoEmoji Font", options=list(available_fonts.keys()))
+        selected_font_name = st.selectbox("Select Font", options=list(available_fonts.keys()), index=0 if "NotoColorEmoji-Regular" in available_fonts else 0)
         selected_font_path = available_fonts[selected_font_name]
         base_font_size = st.slider("Font Size (points)", min_value=6, max_value=16, value=9, step=1)  # Default to 9
         plain_text_mode = st.checkbox("Render as Plain Text (Preserve Bold Only)", value=False)
+        auto_bold_numbers = st.checkbox("Auto-Bold Numbered Lines", value=False)
         num_columns = st.selectbox("Number of Columns", options=[1, 2, 3, 4, 5, 6], index=3)  # Default to 4
 
         # Markdown editor and buttons
@@ -42,35 +44,47 @@ def create_pdf_tab(default_markdown):
         
         st.download_button(label="Save Markdown", data=st.session_state.markdown_content, file_name="deities_guide.md", mime="text/markdown")
 
-    # Register font
+    # Register the selected font with ReportLab
+    # Note: Fonts must be TrueType (.ttf) for ReportLab compatibility.
     pdfmetrics.registerFont(TTFont(selected_font_name, selected_font_path))
 
-    # Emoji font application
+    # Emoji font application function
+    # This function handles Unicode emojis in the text, applying the selected font to them.
+    # We use a regex to match emoji ranges (Unicode blocks like Miscellaneous Symbols, Emoticons, etc.).
+    # To avoid multi-character emoji issues (e.g., base char + variation selector U+FE0F), we limit to the first character.
+    # Noto Color Emoji (NotoColorEmoji-Regular.ttf) is recommended as it supports a full range of modern emojis,
+    # including color rendering, though ReportLab renders in black-and-white by default.
     def apply_emoji_font(text, emoji_font):
         emoji_pattern = re.compile(
-            r"([\U0001F300-\U0001F5FF"
-            r"\U0001F600-\U0001F64F"
-            r"\U0001F680-\U0001F6FF"
-            r"\U0001F700-\U0001F77F"
-            r"\U0001F780-\U0001F7FF"
-            r"\U0001F800-\U0001F8FF"
-            r"\U0001F900-\U0001F9FF"
-            r"\U0001FA00-\U0001FA6F"
-            r"\U0001FA70-\U0001FAFF"
-            r"\u2600-\u26FF"
-            r"\u2700-\u27BF]+)"
+            r"([\U0001F300-\U0001F5FF"  # Miscellaneous Symbols and Pictographs
+            r"\U0001F600-\U0001F64F"  # Emoticons
+            r"\U0001F680-\U0001F6FF"  # Transport and Map Symbols
+            r"\U0001F700-\U0001F77F"  # Alchemical Symbols
+            r"\U0001F780-\U0001F7FF"  # Geometric Shapes Extended
+            r"\U0001F800-\U0001F8FF"  # Supplemental Arrows-C
+            r"\U0001F900-\U0001F9FF"  # Supplemental Symbols and Pictographs
+            r"\U0001FA00-\U0001FA6F"  # Chess Symbols
+            r"\U0001FA70-\U0001FAFF"  # Symbols and Pictographs Extended-A
+            r"\u2600-\u26FF"          # Miscellaneous Symbols (e.g., ⚡ U+26A1)
+            r"\u2700-\u27BF]+"        # Dingbats (e.g., ✝ U+271D)
+            r")"
         )
         def replace_emoji(match):
             emoji = match.group(1)
+            # Limit to first character to avoid rendering issues with multi-codepoint emojis
+            # Example: ✝️ (U+271D U+FE0F) becomes just ✝ (U+271D), dropping the variation selector
             if len(emoji) > 1:
                 emoji = emoji[0]
             return f'<font face="{emoji_font}">{emoji}</font>'
         return emoji_pattern.sub(replace_emoji, text)
 
-    # Markdown to PDF content
-    def markdown_to_pdf_content(markdown_text, plain_text_mode):
+    # Convert markdown to PDF content
+    # This function processes markdown lines, filtering out headers and applying bolding rules.
+    # If auto_bold_numbers is True, lines starting with "number. " (e.g., "1. ") are bolded.
+    def markdown_to_pdf_content(markdown_text, plain_text_mode, auto_bold_numbers):
         lines = markdown_text.strip().split('\n')
         pdf_content = []
+        number_pattern = re.compile(r'^\d+\.\s')
 
         if plain_text_mode:
             for line in lines:
@@ -88,6 +102,8 @@ def create_pdf_tab(default_markdown):
                 if line.startswith('## ') or line.startswith('### '):
                     text = line.replace('## ', '').replace('### ', '').strip()
                     pdf_content.append(f"<b>{text}</b>")
+                elif auto_bold_numbers and number_pattern.match(line):
+                    pdf_content.append(f"<b>{line}</b>")
                 else:
                     pdf_content.append(line.strip())
 
@@ -95,17 +111,20 @@ def create_pdf_tab(default_markdown):
         return pdf_content, total_lines
 
     # Create PDF
-    def create_pdf(markdown_text, base_font_size, plain_text_mode, num_columns):
+    # This function builds a PDF using ReportLab, arranging content in columns.
+    # We use a double A4 landscape layout (A4 width * 2) for wide content.
+    # Spacers are added before numbered sections for visual separation.
+    # Paragraph styles define font sizes and bolding, with emojis rendered via the selected font.
+    def create_pdf(markdown_text, base_font_size, plain_text_mode, num_columns, auto_bold_numbers):
         buffer = io.BytesIO()
-        # Double A4 page: A4 width * 2 (landscape)
         page_width = A4[0] * 2
         page_height = A4[1]
         doc = SimpleDocTemplate(buffer, pagesize=(page_width, page_height), leftMargin=36, rightMargin=36, topMargin=36, bottomMargin=36)
         styles = getSampleStyleSheet()
         story = []
         spacer_height = 10
-        section_spacer_height = 15  # Extra spacing before numbered sections
-        pdf_content, total_lines = markdown_to_pdf_content(markdown_text, plain_text_mode)
+        section_spacer_height = 15
+        pdf_content, total_lines = markdown_to_pdf_content(markdown_text, plain_text_mode, auto_bold_numbers)
 
         item_font_size = base_font_size
         section_font_size = base_font_size * 1.1
@@ -125,12 +144,9 @@ def create_pdf_tab(default_markdown):
         current_line_count = 0
         current_column = 0
 
-        # Regex to detect numbered sections (e.g., "1. ", "2. ")
         number_pattern = re.compile(r'^\d+\.\s')
-
         for i, item in enumerate(pdf_content):
-            # Add extra spacing before numbered sections (but not the first one)
-            if i > 0 and number_pattern.match(item):
+            if i > 0 and number_pattern.match(item.replace('<b>', '').replace('</b>', '')):
                 columns[current_column].append(Spacer(1, section_spacer_height))
             
             if current_line_count >= lines_per_column and current_column < num_columns - 1:
@@ -170,7 +186,8 @@ def create_pdf_tab(default_markdown):
         buffer.seek(0)
         return buffer.getvalue()
 
-    # PDF to image
+    # Convert PDF to image for preview
+    # Uses PyMuPDF (fitz) to render PDF pages as images for Streamlit display.
     def pdf_to_image(pdf_bytes):
         try:
             doc = fitz.open(stream=pdf_bytes, filetype="pdf")
@@ -185,9 +202,9 @@ def create_pdf_tab(default_markdown):
             st.error(f"Failed to render PDF preview: {e}")
             return None
 
-    # Main logic (PDF generation and preview only)
+    # Main logic
     with st.spinner("Generating PDF..."):
-        pdf_bytes = create_pdf(st.session_state.markdown_content, base_font_size, plain_text_mode, num_columns)
+        pdf_bytes = create_pdf(st.session_state.markdown_content, base_font_size, plain_text_mode, num_columns, auto_bold_numbers)
 
     with st.container():
         pdf_images = pdf_to_image(pdf_bytes)
@@ -197,7 +214,6 @@ def create_pdf_tab(default_markdown):
         else:
             st.info("Download the PDF to view it locally.")
 
-    # "Download PDF" in sidebar
     with st.sidebar:
         st.download_button(label="Download PDF", data=pdf_bytes, file_name="deities_guide.pdf", mime="application/pdf")
 
@@ -229,7 +245,7 @@ default_markdown = """# Deities Guide: Mythology and Moral Lessons 🌟
 
 5. 🌠 Ascension and Signs
    - Paths: Birth, deeds, revelation, as with Jesus and Arjuna.
-   - Signs: Miracles and prophecies, like those in the Quran and Gita.
+   - Signs: Miracles and prophecies, like those in the Quran and Gita Kobe.
    - Morals: Obedience and devotion shape destiny.
 
 6. 🎲 Storytelling and Games