123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288 |
- import functools
- import re
- import string
- import typing as t
- if t.TYPE_CHECKING:
- import typing_extensions as te
- class HasHTML(te.Protocol):
- def __html__(self) -> str:
- pass
- __version__ = "2.0.1"
- _striptags_re = re.compile(r"(<!--.*?-->|<[^>]*>)")
- def _simple_escaping_wrapper(name: str) -> t.Callable[..., "Markup"]:
- orig = getattr(str, name)
- @functools.wraps(orig)
- def wrapped(self: "Markup", *args: t.Any, **kwargs: t.Any) -> "Markup":
- args = _escape_argspec(list(args), enumerate(args), self.escape) # type: ignore
- _escape_argspec(kwargs, kwargs.items(), self.escape)
- return self.__class__(orig(self, *args, **kwargs))
- return wrapped
- class Markup(str):
- """A string that is ready to be safely inserted into an HTML or XML
- document, either because it was escaped or because it was marked
- safe.
- Passing an object to the constructor converts it to text and wraps
- it to mark it safe without escaping. To escape the text, use the
- :meth:`escape` class method instead.
- >>> Markup("Hello, <em>World</em>!")
- Markup('Hello, <em>World</em>!')
- >>> Markup(42)
- Markup('42')
- >>> Markup.escape("Hello, <em>World</em>!")
- Markup('Hello <em>World</em>!')
- This implements the ``__html__()`` interface that some frameworks
- use. Passing an object that implements ``__html__()`` will wrap the
- output of that method, marking it safe.
- >>> class Foo:
- ... def __html__(self):
- ... return '<a href="/foo">foo</a>'
- ...
- >>> Markup(Foo())
- Markup('<a href="/foo">foo</a>')
- This is a subclass of :class:`str`. It has the same methods, but
- escapes their arguments and returns a ``Markup`` instance.
- >>> Markup("<em>%s</em>") % ("foo & bar",)
- Markup('<em>foo & bar</em>')
- >>> Markup("<em>Hello</em> ") + "<foo>"
- Markup('<em>Hello</em> <foo>')
- """
- __slots__ = ()
- def __new__(
- cls, base: t.Any = "", encoding: t.Optional[str] = None, errors: str = "strict"
- ) -> "Markup":
- if hasattr(base, "__html__"):
- base = base.__html__()
- if encoding is None:
- return super().__new__(cls, base)
- return super().__new__(cls, base, encoding, errors)
- def __html__(self) -> "Markup":
- return self
- def __add__(self, other: t.Union[str, "HasHTML"]) -> "Markup":
- if isinstance(other, str) or hasattr(other, "__html__"):
- return self.__class__(super().__add__(self.escape(other)))
- return NotImplemented
- def __radd__(self, other: t.Union[str, "HasHTML"]) -> "Markup":
- if isinstance(other, str) or hasattr(other, "__html__"):
- return self.escape(other).__add__(self)
- return NotImplemented
- def __mul__(self, num: int) -> "Markup":
- if isinstance(num, int):
- return self.__class__(super().__mul__(num))
- return NotImplemented # type: ignore
- __rmul__ = __mul__
- def __mod__(self, arg: t.Any) -> "Markup":
- if isinstance(arg, tuple):
- arg = tuple(_MarkupEscapeHelper(x, self.escape) for x in arg)
- else:
- arg = _MarkupEscapeHelper(arg, self.escape)
- return self.__class__(super().__mod__(arg))
- def __repr__(self) -> str:
- return f"{self.__class__.__name__}({super().__repr__()})"
- def join(self, seq: t.Iterable[t.Union[str, "HasHTML"]]) -> "Markup":
- return self.__class__(super().join(map(self.escape, seq)))
- join.__doc__ = str.join.__doc__
- def split( # type: ignore
- self, sep: t.Optional[str] = None, maxsplit: int = -1
- ) -> t.List["Markup"]:
- return [self.__class__(v) for v in super().split(sep, maxsplit)]
- split.__doc__ = str.split.__doc__
- def rsplit( # type: ignore
- self, sep: t.Optional[str] = None, maxsplit: int = -1
- ) -> t.List["Markup"]:
- return [self.__class__(v) for v in super().rsplit(sep, maxsplit)]
- rsplit.__doc__ = str.rsplit.__doc__
- def splitlines(self, keepends: bool = False) -> t.List["Markup"]: # type: ignore
- return [self.__class__(v) for v in super().splitlines(keepends)]
- splitlines.__doc__ = str.splitlines.__doc__
- def unescape(self) -> str:
- """Convert escaped markup back into a text string. This replaces
- HTML entities with the characters they represent.
- >>> Markup("Main » <em>About</em>").unescape()
- 'Main » <em>About</em>'
- """
- from html import unescape
- return unescape(str(self))
- def striptags(self) -> str:
- """:meth:`unescape` the markup, remove tags, and normalize
- whitespace to single spaces.
- >>> Markup("Main »\t<em>About</em>").striptags()
- 'Main » About'
- """
- stripped = " ".join(_striptags_re.sub("", self).split())
- return Markup(stripped).unescape()
- @classmethod
- def escape(cls, s: t.Any) -> "Markup":
- """Escape a string. Calls :func:`escape` and ensures that for
- subclasses the correct type is returned.
- """
- rv = escape(s)
- if rv.__class__ is not cls:
- return cls(rv)
- return rv
- for method in (
- "__getitem__",
- "capitalize",
- "title",
- "lower",
- "upper",
- "replace",
- "ljust",
- "rjust",
- "lstrip",
- "rstrip",
- "center",
- "strip",
- "translate",
- "expandtabs",
- "swapcase",
- "zfill",
- ):
- locals()[method] = _simple_escaping_wrapper(method)
- del method
- def partition(self, sep: str) -> t.Tuple["Markup", "Markup", "Markup"]:
- l, s, r = super().partition(self.escape(sep))
- cls = self.__class__
- return cls(l), cls(s), cls(r)
- def rpartition(self, sep: str) -> t.Tuple["Markup", "Markup", "Markup"]:
- l, s, r = super().rpartition(self.escape(sep))
- cls = self.__class__
- return cls(l), cls(s), cls(r)
- def format(self, *args: t.Any, **kwargs: t.Any) -> "Markup":
- formatter = EscapeFormatter(self.escape)
- return self.__class__(formatter.vformat(self, args, kwargs))
- def __html_format__(self, format_spec: str) -> "Markup":
- if format_spec:
- raise ValueError("Unsupported format specification for Markup.")
- return self
- class EscapeFormatter(string.Formatter):
- __slots__ = ("escape",)
- def __init__(self, escape: t.Callable[[t.Any], Markup]) -> None:
- self.escape = escape
- super().__init__()
- def format_field(self, value: t.Any, format_spec: str) -> str:
- if hasattr(value, "__html_format__"):
- rv = value.__html_format__(format_spec)
- elif hasattr(value, "__html__"):
- if format_spec:
- raise ValueError(
- f"Format specifier {format_spec} given, but {type(value)} does not"
- " define __html_format__. A class that defines __html__ must define"
- " __html_format__ to work with format specifiers."
- )
- rv = value.__html__()
- else:
- # We need to make sure the format spec is str here as
- # otherwise the wrong callback methods are invoked.
- rv = string.Formatter.format_field(self, value, str(format_spec))
- return str(self.escape(rv))
- _ListOrDict = t.TypeVar("_ListOrDict", list, dict)
- def _escape_argspec(
- obj: _ListOrDict, iterable: t.Iterable[t.Any], escape: t.Callable[[t.Any], Markup]
- ) -> _ListOrDict:
- """Helper for various string-wrapped functions."""
- for key, value in iterable:
- if isinstance(value, str) or hasattr(value, "__html__"):
- obj[key] = escape(value)
- return obj
- class _MarkupEscapeHelper:
- """Helper for :meth:`Markup.__mod__`."""
- __slots__ = ("obj", "escape")
- def __init__(self, obj: t.Any, escape: t.Callable[[t.Any], Markup]) -> None:
- self.obj = obj
- self.escape = escape
- def __getitem__(self, item: t.Any) -> "_MarkupEscapeHelper":
- return _MarkupEscapeHelper(self.obj[item], self.escape)
- def __str__(self) -> str:
- return str(self.escape(self.obj))
- def __repr__(self) -> str:
- return str(self.escape(repr(self.obj)))
- def __int__(self) -> int:
- return int(self.obj)
- def __float__(self) -> float:
- return float(self.obj)
- # circular import
- try:
- from ._speedups import escape as escape
- from ._speedups import escape_silent as escape_silent
- from ._speedups import soft_str as soft_str
- from ._speedups import soft_unicode
- except ImportError:
- from ._native import escape as escape
- from ._native import escape_silent as escape_silent # noqa: F401
- from ._native import soft_str as soft_str # noqa: F401
- from ._native import soft_unicode # noqa: F401
|