import io import re import os import glob import asyncio import hashlib import unicodedata import streamlit as st from PIL import Image import fitz import edge_tts from reportlab.lib.pagesizes import A4 from reportlab.platypus import SimpleDocTemplate, Paragraph, Spacer, Table, TableStyle from reportlab.lib.styles import getSampleStyleSheet, ParagraphStyle from reportlab.lib import colors from reportlab.pdfbase import pdfmetrics from reportlab.pdfbase.ttfonts import TTFont from datetime import datetime import pytz st.set_page_config(layout="wide", initial_sidebar_state="collapsed") # πŸ•’ Time flies when you're having function calls def get_timestamp_prefix(): """πŸ•°οΈ Gets a timestamp that's more reliable than your coworker's ETA estimates""" central = pytz.timezone("US/Central") now = datetime.now(central) # Format: three-letter day, MMDD, HHMM + AM/PM (all uppercase) return now.strftime("%a %m%d %I%M%p").upper() # 🧹 Because text needs a bath before being spoken def clean_for_speech(text): """🧼 Scrubs your text cleaner than your bathroom will ever be""" # Remove hash marks text = text.replace("#", "") # Remove emojis using a regex pattern that covers a wide range 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]+", flags=re.UNICODE) text = emoji_pattern.sub('', text) return text # 🎀 Making robots talk so you don't have to async def generate_audio(text, voice, filename): """πŸ”Š Turn text into speech, because reading is so last century""" communicate = edge_tts.Communicate(text, voice) await communicate.save(filename) return filename # πŸ”— Detecting links like a digital bloodhound def detect_and_convert_links(text): """πŸ•ΈοΈ 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'\1', 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 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(' 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'{url}') start_idx = match.end() return ''.join(result) # 🎭 Making emojis wear the right font costume def apply_emoji_font(text, emoji_font): """πŸ¦„ Because emojis deserve their own font fashion show""" # First handle links - temporarily replace them with placeholders link_pattern = re.compile(r'(.*?)') 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'(.*?)', lambda m: f'###BOLD_START###{m.group(1)}###BOLD_END###', text) # Apply emoji font replacement 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"\U0001FAD0-\U0001FAD9" # additional range if needed r"\U0001FA00-\U0001FA6F" r"\U0001FA70-\U0001FAFF" r"\u2600-\u26FF" r"\u2700-\u27BF]+)" ) def replace_emoji(match): emoji = match.group(1) emoji = unicodedata.normalize('NFC', emoji) return f'{emoji}' segments = [] last_pos = 0 for match in emoji_pattern.finditer(text): start, end = match.span() if last_pos < start: segments.append(f'{text[last_pos:start]}') segments.append(replace_emoji(match)) last_pos = end if last_pos < len(text): segments.append(f'{text[last_pos:]}') combined_text = ''.join(segments) # Restore bold tags combined_text = combined_text.replace('###BOLD_START###', '') combined_text = combined_text.replace('###BOLD_END###', '') # 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(' before.rfind(''): # Close font tag before link, reopen after link_html = f'{label}' combined_text = before + link_html + after else: # Simple replacement combined_text = before + f'{label}' + after return combined_text # πŸ“ Converting markdown to PDF content, because PDFs never go out of style def markdown_to_pdf_content(markdown_text, render_with_bold, auto_bold_numbers): """πŸ“‹ Transforms your disorganized thoughts into structured PDF content""" lines = markdown_text.strip().split('\n') pdf_content = [] number_pattern = re.compile(r'^\d+\.\s') for line in lines: 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 if render_with_bold: line = re.sub(r'\*\*(.*?)\*\*', r'\1', line) if auto_bold_numbers and number_pattern.match(line): # Only apply bold if not already entirely bold if not (line.startswith("") and line.endswith("")): # If there's already some bold formatting inside, we need to handle carefully if "" in line and "" in line: # Complex case - for simplicity, just make the whole line bold # but remove any existing bold tags first line = re.sub(r'', '', line) line = f"{line}" else: line = f"{line}" pdf_content.append(line) total_lines = len(pdf_content) return pdf_content, total_lines # πŸ—οΈ Building PDFs like it's your second job def create_pdf(markdown_text, base_font_size, render_with_bold, auto_bold_numbers, enlarge_numbered, num_columns): """πŸ”¨ Constructs a PDF with the precision of a sleep-deprived architect""" buffer = io.BytesIO() 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() 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, 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, 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, linkUnderline=True # Enable underline for links ) # Register fonts try: available_font_files = glob.glob("*.ttf") if not available_font_files: st.error("No .ttf font files found in the current directory.") return selected_font_path = None for f in available_font_files: if "NotoEmoji-Bold" in f: selected_font_path = f break if selected_font_path: pdfmetrics.registerFont(TTFont("NotoEmoji-Bold", selected_font_path)) pdfmetrics.registerFont(TTFont("DejaVuSans", "DejaVuSans.ttf")) 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: if isinstance(item, str) and item.startswith("") and item.endswith(""): content = item[3:-4].strip() if number_pattern.match(content): column_cells[col_idx].append(Paragraph(apply_emoji_font(content, "NotoEmoji-Bold"), numbered_bold_style)) else: 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') table.setStyle(TableStyle([ ('VALIGN', (0, 0), (-1, -1), 'TOP'), ('ALIGN', (0, 0), (-1, -1), 'LEFT'), ('BACKGROUND', (0, 0), (-1, -1), colors.white), ('GRID', (0, 0), (-1, -1), 0, colors.white), ('LINEAFTER', (0, 0), (num_columns-1, -1), 0.5, colors.grey), ('LEFTPADDING', (0, 0), (-1, -1), 2), ('RIGHTPADDING', (0, 0), (-1, -1), 2), ('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) return buffer.getvalue() # πŸ–ΌοΈ Converting PDFs to images, because we can't leave well enough alone def pdf_to_image(pdf_bytes): """πŸ”Ž Turns your PDF into pictures because some people just want to see the world rendered""" try: doc = fitz.open(stream=pdf_bytes, filetype="pdf") images = [] for page in doc: pix = page.get_pixmap(matrix=fitz.Matrix(2.0, 2.0)) img = Image.frombytes("RGB", [pix.width, pix.height], pix.samples) images.append(img) doc.close() return images except Exception as e: st.error(f"Failed to render PDF preview: {e}") return None # -- Markdown File Selection -- md_files = [f for f in glob.glob("*.md") if os.path.basename(f) != "README.md"] md_options = [os.path.splitext(os.path.basename(f))[0] for f in md_files] # πŸŽͺ The main Streamlit show begins here with st.sidebar: st.markdown("### PDF Options") if md_options: selected_md = st.selectbox("Select Markdown File", options=md_options, index=0) with open(f"{selected_md}.md", "r", encoding="utf-8") as f: st.session_state.markdown_content = f.read() else: st.warning("No markdown file found. Please add one to your folder.") selected_md = None st.session_state.markdown_content = "" available_font_files = {os.path.splitext(os.path.basename(f))[0]: f for f in glob.glob("*.ttf")} selected_font_name = st.selectbox("Select Emoji Font", options=list(available_font_files.keys()), index=list(available_font_files.keys()).index("NotoEmoji-Bold") if "NotoEmoji-Bold" in available_font_files else 0) base_font_size = st.slider("Font Size (points)", min_value=6, max_value=16, value=8, step=1) render_with_bold = st.checkbox("Render with Bold Formatting (remove ** markers)", value=True, key="render_with_bold") auto_bold_numbers = st.checkbox("Auto Bold Numbered Lines", value=True, key="auto_bold_numbers") enlarge_numbered = st.checkbox("Enlarge Font Size for Numbered Lines", value=True, key="enlarge_numbered") num_columns = st.selectbox("Number of Columns", options=[1, 2, 3, 4, 5, 6], index=3) # Use the file's content for editing edited_markdown = st.text_area("Modify the markdown content below:", value=st.session_state.markdown_content, height=300, key=f"markdown_{selected_md}_{selected_font_name}_{num_columns}") if st.button("Update PDF"): st.session_state.markdown_content = edited_markdown if selected_md: with open(f"{selected_md}.md", "w", encoding="utf-8") as f: f.write(edited_markdown) st.rerun() # Create a timestamp prefix for file naming prefix = get_timestamp_prefix() # Download button for Markdown with new naming convention and double emoji label st.download_button( label="πŸ’ΎπŸ“ Save Markdown", data=st.session_state.markdown_content, file_name=f"{prefix} {selected_md}.md" if selected_md else f"{prefix} default.md", mime="text/markdown" ) st.markdown("### Text-to-Speech") VOICES = ["en-US-AriaNeural", "en-US-JennyNeural", "en-GB-SoniaNeural", "en-US-GuyNeural", "en-US-AnaNeural"] selected_voice = st.selectbox("Select Voice for TTS", options=VOICES, index=0) if st.button("Generate Audio"): # Clean markdown input for speech generation cleaned_text = clean_for_speech(st.session_state.markdown_content) # Create a filename for the audio file using the timestamp, markdown name, and selected voice audio_filename = f"{prefix} {selected_md} {selected_voice}.mp3" if selected_md else f"{prefix} default {selected_voice}.mp3" audio_file = asyncio.run(generate_audio(cleaned_text, selected_voice, audio_filename)) st.audio(audio_file) with open(audio_file, "rb") as f: audio_bytes = f.read() st.download_button( label="πŸ’ΎπŸ”Š Save Audio", data=audio_bytes, file_name=audio_filename, mime="audio/mpeg" ) # πŸš€ Generating the PDF with more complexity than a rocket launch with st.spinner("Generating PDF..."): pdf_bytes = create_pdf(st.session_state.markdown_content, base_font_size, render_with_bold, auto_bold_numbers, enlarge_numbered, num_columns) # πŸ“Ί Displaying the preview, because everyone loves to window shop with st.container(): pdf_images = pdf_to_image(pdf_bytes) if pdf_images: for img in pdf_images: st.image(img, use_container_width=True) else: st.info("Download the PDF to view it locally.") # πŸ’Ύ Last chance to save your masterpiece before it's gone forever with st.sidebar: st.download_button( label="πŸ’ΎπŸ“„ Save PDF", data=pdf_bytes, file_name=f"{prefix} {selected_md}.pdf" if selected_md else f"{prefix} output.pdf", mime="application/pdf" )