Class MarkdownImageUtil

docreader/parser/markdown_parser.py:227–413

Utility class for handling images in Markdown. This class provides functionality to: - Extract base64-encoded images from Markdown - Extract image paths from Markdown - Replace image paths with new URLs - Convert base64 images to binary format Supported formats: - Base6

class MarkdownImageUtil:
    """Utility class for handling images in Markdown.

    This class provides functionality to:
    - Extract base64-encoded images from Markdown
    - Extract image paths from Markdown
    - Replace image paths with new URLs
    - Convert base64 images to binary format

    Supported formats:
    - Base64 embedded images: ![alt](data:image/png;base64,iVBORw0...)
    - Regular image links: ![alt](path/to/image.png)
    """

    def __init__(self):
        # Pattern to match base64 embedded images
        # Captures: (1) alt text, (2) image format, (3) base64 data
        # Alt text uses .*? (non-greedy) to allow literal ] (e.g. Windows paths).
        # MIME subtype uses [^;]+ to handle types with hyphens like x-emf.
        self.b64_pattern = re.compile(
            r"!\[(.*?)\]\(data:image/([^;]+);base64,([^\)]+)\)"
        )
        # Pattern to match regular image syntax (alt text allows ])
        self.image_pattern = re.compile(r"!\[(.*?)\]\(([^)]+)\)")
        # Pattern for replacing image paths
        self.replace_pattern = re.compile(r"!\[(.*?)\]\(([^)]+)\)")

    def extract_image(
        self,
        content: str,
        path_prefix: Optional[str] = None,
        replace: bool = True,
    ) -> Tuple[str, List[str]]:
        """Extract image paths from Markdown content.

        Args:
            content: Markdown text containing images
            path_prefix: Optional prefix to add to image paths
            replace: Whether to replace image syntax in content

        Returns:
            Tuple of (processed_text, list_of_image_paths)

        Example:
            >>> util = MarkdownImageUtil()
            >>> text, images = util.extract_image("![logo](img/logo.png)")
            >>> print(images)
            ['img/logo.png']
        """
        # List to store extracted image paths
        images: List[str] = []

        def repl(match: Match[str]) -> str:
            """Replacement function for each image match."""
            title = match.group(1)  # Alt text
            image_path = match.group(2)  # Image path

            # Add prefix if specified