audiobook-maker-pro-v4.2/routes/public_routes.py

# routes/public_routes.py - Public (No Auth) Routes for Published Audiobooks

import re
import json
from flask import Blueprint, jsonify, send_from_directory, abort

from db import get_db

public_bp = Blueprint('public', __name__)


# ============================================
# Helpers
# ============================================

_CONTROL_CHAR_RE = re.compile(r'[\x00-\x08\x0B\x0C\x0E-\x1F\x7F]')


def clean_str(s):
    if s is None:
        return ''
    if not isinstance(s, str):
        s = str(s)
    return _CONTROL_CHAR_RE.sub('', s)


def clean_transcription(transcription):
    if isinstance(transcription, list):
        for t in transcription:
            if isinstance(t, dict) and 'word' in t:
                t['word'] = clean_str(t.get('word', ''))
    return transcription


# ============================================
# Routes
# ============================================

@public_bp.route('/home')
def public_home():
    """Public homepage - Bookcase view of published audiobooks."""
    return send_from_directory('templates', 'public_home.html')


@public_bp.route('/read/<int:project_id>')
def public_reader(project_id):
    """Public reader page for a published audiobook."""
    db = get_db()
    cursor = db.cursor()
    cursor.execute('SELECT id, is_published FROM projects WHERE id = ?', (project_id,))
    project = cursor.fetchone()
    
    if not project or not project['is_published']:
        abort(404)
    
    # Increment view count
    cursor.execute('UPDATE projects SET view_count = view_count + 1 WHERE id = ?', (project_id,))
    db.commit()
    
    return send_from_directory('templates', 'public_reader.html')


@public_bp.route('/api/public/books', methods=['GET'])
def list_published_books():
    """List all published audiobooks (no auth required)."""
    db = get_db()
    cursor = db.cursor()
    
    cursor.execute('''
        SELECT p.id, p.name, p.description, p.author, p.category,
               p.thumbnail_data, p.thumbnail_format, p.published_at,
               p.view_count, p.created_at,
               (SELECT COUNT(*) FROM chapters WHERE project_id = p.id) as chapter_count
        FROM projects p
        WHERE p.is_published = 1
        ORDER BY p.published_at DESC
    ''')
    
    books = []
    for row in cursor.fetchall():
        books.append({
            'id': row['id'],
            'name': row['name'],
            'description': row['description'] or '',
            'author': row['author'] or '',
            'category': row['category'] or '',
            'thumbnail_data': row['thumbnail_data'],
            'thumbnail_format': row['thumbnail_format'] or 'png',
            'published_at': row['published_at'],
            'view_count': row['view_count'] or 0,
            'chapter_count': row['chapter_count']
        })
    
    return jsonify({'books': books})


@public_bp.route('/api/public/books/<int:project_id>', methods=['GET'])
def get_published_book(project_id):
    """
    Get book metadata WITHOUT audio_data.
    Audio is loaded lazily via /api/public/books/<id>/audio/<block_id>.
    This keeps the response small (<1 MB) and avoids proxy truncation issues.
    """
    db = get_db()
    cursor = db.cursor()
    
    cursor.execute('''
        SELECT * FROM projects WHERE id = ? AND is_published = 1
    ''', (project_id,))
    project = cursor.fetchone()
    
    if not project:
        return jsonify({'error': 'Book not found or not published'}), 404
    
    cursor.execute('''
        SELECT * FROM chapters WHERE project_id = ? ORDER BY chapter_number
    ''', (project_id,))
    chapters = cursor.fetchall()
    
    chapters_data = []
    for chapter in chapters:
        cursor.execute('''
            SELECT id, block_order, block_type, content, audio_format, transcription,
                   (audio_data IS NOT NULL AND audio_data != '') as has_audio
            FROM markdown_blocks WHERE chapter_id = ? ORDER BY block_order
        ''', (chapter['id'],))
        blocks = cursor.fetchall()
        
        blocks_data = []
        for block in blocks:
            cursor.execute('''
                SELECT * FROM block_images WHERE block_id = ? ORDER BY id
            ''', (block['id'],))
            images = cursor.fetchall()
            
            transcription = []
            if block['transcription']:
                try:
                    transcription = json.loads(block['transcription'])
                    transcription = clean_transcription(transcription)
                except (json.JSONDecodeError, TypeError):
                    transcription = []
            
            blocks_data.append({
                'id': block['id'],
                'block_order': block['block_order'],
                'block_type': clean_str(block['block_type']),
                'content': clean_str(block['content']),
                'audio_data': '',  # Empty here; loaded lazily by frontend
                'audio_format': clean_str(block['audio_format']) or 'mp3',
                'has_audio': bool(block['has_audio']),
                'transcription': transcription,
                'images': [{
                    'data': clean_str(img['image_data']),
                    'format': clean_str(img['image_format']) or 'png',
                    'alt_text': clean_str(img['alt_text']),
                    'position': clean_str(img['position']) or 'before'
                } for img in images]
            })
        
        chapters_data.append({
            'id': chapter['id'],
            'chapter_number': chapter['chapter_number'],
            'title': clean_str(chapter['title']),
            'blocks': blocks_data
        })
    
    return jsonify({
        'id': project['id'],
        'name': clean_str(project['name']),
        'description': clean_str(project['description']) if project['description'] else '',
        'author': clean_str(project['author']) if project['author'] else '',
        'thumbnail_data': project['thumbnail_data'],
        'thumbnail_format': project['thumbnail_format'] or 'png',
        'chapters': chapters_data
    })


@public_bp.route('/api/public/books/<int:project_id>/audio/<int:block_id>', methods=['GET'])
def get_public_block_audio(project_id, block_id):
    """
    Return audio_data (base64) for a single block in a published book.
    No auth required since the book is published publicly.
    """
    db = get_db()
    cursor = db.cursor()
    
    # Verify project is published
    cursor.execute('SELECT is_published FROM projects WHERE id = ?', (project_id,))
    project = cursor.fetchone()
    if not project or not project['is_published']:
        return jsonify({'error': 'Book not found or not published'}), 404
    
    cursor.execute('''
        SELECT mb.audio_data, mb.audio_format
        FROM markdown_blocks mb
        JOIN chapters c ON mb.chapter_id = c.id
        WHERE mb.id = ? AND c.project_id = ?
    ''', (block_id, project_id))
    row = cursor.fetchone()
    
    if not row:
        return jsonify({'error': 'Block not found'}), 404
    
    if not row['audio_data']:
        return jsonify({'audio_data': '', 'audio_format': row['audio_format'] or 'mp3'})
    
    return jsonify({
        'audio_data': clean_str(row['audio_data']),
        'audio_format': clean_str(row['audio_format']) or 'mp3'
    })