Class Parser

python/markupever/parser.py:7–120

class Parser:
    __slots__ = ("__raw", "__state")

    def __init__(
        self,
        options: typing.Union[
            _rustlib.HtmlOptions,
            _rustlib.XmlOptions,
            typing.Literal["html"],
            typing.Literal["xml"],
        ] = "html",
    ):
        """
        An HTML/XML parser, ready to receive unicode input.

        This is very easy to use and allows you to stream input using `.process()` method; By this way
        you are don't worry about memory usages of huge inputs.

        for `options`, If your input is a HTML document, pass a `HtmlOptions`;
        If your input is a XML document, pass `XmlOptions`.
        """
        if isinstance(options, str):
            if options == "html":
                options = _rustlib.HtmlOptions()
            elif options == "xml":
                options = _rustlib.XmlOptions()
            else:
                raise ValueError(f"invalid parser options: {options!r}")

        self.__raw = _rustlib.Parser(options)

        # 0 - processing
        # 1 - finished
        # 2 - converted
        self.__state = 0

    def writable(self) -> bool:
        """
        Shorthand for `not Parser.is_finished`.

        This function exists to make `Parser` like a `BytesIO` and `StringIO`.
        You can pass the `Parser` to each function which needs a writable buffer or IO.
        """
        return not self.is_finished

    def write(self, content: typing.Union[str, bytes]) -> int:
        """
        Same as `Parser.process`.

        This function exists to make `Parser` like a `BytesIO` and `StringIO`.
        You can pass the `Parser` to each function which needs a writable buffer or IO.
        """
        self.__raw.process(content)
        return len(content)

    def process(self, content: typing.Union[str, bytes]) -> "Parser":
        """
        Processes an input.