app.py

import io
import re
import os
import glob
import asyncio
import hashlib
import base64
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

st.set_page_config(layout="wide", initial_sidebar_state="collapsed")

def clean_text_for_tts(text):
    # Remove asterisks, pound signs, and emojis from text for audio generation.
    text = re.sub(r'[#*]', '', text)
    emoji_pattern = re.compile("["
        u"\U0001F600-\U0001F64F"  # emoticons
        u"\U0001F300-\U0001F5FF"  # symbols & pictographs
        u"\U0001F680-\U0001F6FF"  # transport & map symbols
        u"\U0001F1E0-\U0001F1FF"  # flags
                           "]+", flags=re.UNICODE)
    text = emoji_pattern.sub(r'', text)
    return text.strip()

def get_file_title_from_markdown(markdown_text):
    # Extract first sizable line (starting with '#' and non-empty) and clean it for use as a file name.
    for line in markdown_text.splitlines():
        if line.strip() and line.lstrip().startswith("#"):
            title = line.lstrip("#").strip()
            title = re.sub(r'[^A-Za-z0-9 ]+', '', title).strip()
            if title:
                return title.replace(" ", "_")
    return "output"

async def generate_audio(text, voice, markdown_text):
    # Clean the text and generate a file name based on the markdown title.
    cleaned_text = clean_text_for_tts(text)
    title = get_file_title_from_markdown(markdown_text)
    filename = f"{title}.mp3"
    communicate = edge_tts.Communicate(cleaned_text, voice)
    await communicate.save(filename)
    return filename

def get_download_link(file, file_type="mp3"):
    # Generate a base64 download link for a file.
    with open(file, "rb") as f:
        b64 = base64.b64encode(f.read()).decode()
    if file_type == "mp3":
        mime = "audio/mpeg"
    elif file_type == "pdf":
        mime = "application/pdf"
    else:
        mime = "application/octet-stream"
    return f'<a href="data:{mime};base64,{b64}" download="{os.path.basename(file)}">Download {os.path.basename(file)}</a>'

def apply_emoji_font(text, emoji_font):
    # Replace emoji characters with HTML font tags using the specified 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]+)"
    )
    def replace_emoji(match):
        emoji = match.group(1)
        emoji = unicodedata.normalize('NFC', emoji)
        return f'<font face="{emoji_font}">{emoji}</font>'
    segments = []
    last_pos = 0
    for match in emoji_pattern.finditer(text):
        start, end = match.span()
        if last_pos < start:
            segments.append(f'<font face="{emoji_font}">{text[last_pos:start]}</font>')
        segments.append(replace_emoji(match))
        last_pos = end
    if last_pos < len(text):
        segments.append(f'<font face="{emoji_font}">{text[last_pos:]}</font>')
    return ''.join(segments)

def markdown_to_pdf_content(markdown_text, render_with_bold, auto_bold_numbers):
    # Convert markdown text into a list of lines with optional bold formatting.
    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
        if render_with_bold:
            line = re.sub(r'\*\*(.*?)\*\*', r'<b>\1</b>', line)
        if auto_bold_numbers and number_pattern.match(line):
            if not (line.startswith("<b>") and line.endswith("</b>")):
                line = f"<b>{line}</b>"
        pdf_content.append(line)
    total_lines = len(pdf_content)
    return pdf_content, total_lines

def create_pdf(markdown_text, base_font_size, render_with_bold, auto_bold_numbers, enlarge_numbered, num_columns, emoji_font):
    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
    section_spacer_height = 15
    pdf_content, total_lines = markdown_to_pdf_content(markdown_text, render_with_bold, auto_bold_numbers)
    item_style = ParagraphStyle(
        'ItemStyle', parent=styles['Normal'], fontName="DejaVuSans",
        fontSize=base_font_size, leading=base_font_size * 1.15, spaceAfter=1
    )
    bold_style = ParagraphStyle(
        'BoldStyle', parent=styles['Normal'], fontName="NotoEmoji-Bold",
        fontSize=base_font_size, leading=base_font_size * 1.15, spaceAfter=1
    )
    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
    )
    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
    )
    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
    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
    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("<b>") and item.endswith("</b>"):
                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, emoji_font), item_style))
    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)))
    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),
    ]))
    story = [Spacer(1, spacer_height), table]
    doc.build(story)
    buffer.seek(0)
    return buffer.getvalue()

def pdf_to_image(pdf_bytes):
    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

# Auto-detect default markdown file from available .md files.
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]
if md_options:
    if 'markdown_content' not in st.session_state or not st.session_state.markdown_content:
        with open(f"{md_options[0]}.md", "r", encoding="utf-8") as f:
            st.session_state.markdown_content = f.read()
else:
    st.session_state.markdown_content = ""

with st.sidebar:
    st.markdown("### PDF Options")
    selected_md = st.selectbox("Select Markdown File", options=md_options, index=0 if md_options else -1)
    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)
    if md_options and selected_md:
        with open(f"{selected_md}.md", "r", encoding="utf-8") as f:
            st.session_state.markdown_content = f.read()
    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 md_options and selected_md:
            with open(f"{selected_md}.md", "w", encoding="utf-8") as f:
                f.write(edited_markdown)
        st.experimental_rerun()
    st.download_button(label="Save Markdown", data=st.session_state.markdown_content, file_name=f"{selected_md}.md" if selected_md else "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"):
        audio_file = asyncio.run(generate_audio(st.session_state.markdown_content, selected_voice, st.session_state.markdown_content))
        st.audio(audio_file)
        with open(audio_file, "rb") as f:
            audio_bytes = f.read()
        st.download_button("Download Audio", data=audio_bytes, file_name=os.path.basename(audio_file), mime="audio/mpeg")
    if st.button("Save PDF"):
        title = get_file_title_from_markdown(st.session_state.markdown_content)
        pdf_filename = f"{title}.pdf"
        with open(pdf_filename, "wb") as f:
            f.write(pdf_bytes)
        st.success(f"Saved PDF as {pdf_filename}")
        st.experimental_rerun()
    st.markdown("### Saved Audio Files")
    mp3_files = glob.glob("*.mp3")
    for mp3 in mp3_files:
        st.audio(mp3)
        st.markdown(get_download_link(mp3, "mp3"), unsafe_allow_html=True)
    if st.button("Delete All MP3"):
        for mp3 in mp3_files:
            try:
                os.remove(mp3)
            except Exception as e:
                st.error(f"Error deleting {mp3}: {e}")
        st.experimental_rerun()
    st.markdown("### Saved PDF Files")
    pdf_files = glob.glob("*.pdf")
    for pdf in pdf_files:
        st.markdown(get_download_link(pdf, "pdf"), unsafe_allow_html=True)

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, selected_font_name)

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.")

with st.sidebar:
    st.download_button(label="Download PDF", data=pdf_bytes, file_name="output.pdf", mime="application/pdf")