Function process_epub

reader3.py:175–283

(epub_path: str, output_dir: str)

# --- Main Conversion Logic ---

def process_epub(epub_path: str, output_dir: str) -> Book:

    # 1. Load Book
    print(f"Loading {epub_path}...")
    book = epub.read_epub(epub_path)

    # 2. Extract Metadata
    metadata = extract_metadata_robust(book)

    # 3. Prepare Output Directories
    if os.path.exists(output_dir):
        shutil.rmtree(output_dir)
    images_dir = os.path.join(output_dir, 'images')
    os.makedirs(images_dir, exist_ok=True)

    # 4. Extract Images & Build Map
    print("Extracting images...")
    image_map = {} # Key: internal_path, Value: local_relative_path

    for item in book.get_items():
        if item.get_type() == ebooklib.ITEM_IMAGE:
            # Normalize filename
            original_fname = os.path.basename(item.get_name())
            # Sanitize filename for OS
            safe_fname = "".join([c for c in original_fname if c.isalpha() or c.isdigit() or c in '._-']).strip()

            # Save to disk
            local_path = os.path.join(images_dir, safe_fname)
            with open(local_path, 'wb') as f:
                f.write(item.get_content())

            # Map keys: We try both the full internal path and just the basename
            # to be robust against messy HTML src attributes
            rel_path = f"images/{safe_fname}"
            image_map[item.get_name()] = rel_path
            image_map[original_fname] = rel_path

    # 5. Process TOC
    print("Parsing Table of Contents...")
    toc_structure = parse_toc_recursive(book.toc)
    if not toc_structure:
        print("Warning: Empty TOC, building fallback from Spine...")
        toc_structure = get_fallback_toc(book)

    # 6. Process Content (Spine-based to preserve HTML validity)
    print("Processing chapters...")
    spine_chapters = []

    # We iterate over the spine (linear reading order)
    for i, spine_item in enumerate(book.spine):
        item_id, linear = spine_item
        item = book.get_item_with_id(item_id)

        if not item:
            continue

        if item.get_type() == ebooklib.ITEM_DOCUMENT:
            # Raw content