Skip to content
This repository

HTTPS clone URL

Subversion checkout URL

You can clone with HTTPS or Subversion.

Download ZIP

A Python module that implements the jinja2.Markup string

branch: master
README.rst

MarkupSafe

Implements a unicode subclass that supports HTML strings:

>>> from markupsafe import Markup, escape
>>> escape("<script>alert(document.cookie);</script>")
Markup(u'&lt;script&gt;alert(document.cookie);&lt;/script&gt;')
>>> tmpl = Markup("<em>%s</em>")
>>> tmpl % "Peter > Lustig"
Markup(u'<em>Peter &gt; Lustig</em>')

If you want to make an object unicode that is not yet unicode but don't want to lose the taint information, you can use the soft_unicode function. (On Python 3 you can also use soft_str which is a different name for the same function).

>>> from markupsafe import soft_unicode
>>> soft_unicode(42)
u'42'
>>> soft_unicode(Markup('foo'))
Markup(u'foo')

HTML Representations

Objects can customize their HTML markup equivalent by overriding the __html__ function:

>>> class Foo(object):
...  def __html__(self):
...   return '<strong>Nice</strong>'
...
>>> escape(Foo())
Markup(u'<strong>Nice</strong>')
>>> Markup(Foo())
Markup(u'<strong>Nice</strong>')

Silent Escapes

Since MarkupSafe 0.10 there is now also a separate escape function called escape_silent that returns an empty string for None for consistency with other systems that return empty strings for None when escaping (for instance Pylons' webhelpers).

If you also want to use this for the escape method of the Markup object, you can create your own subclass that does that:

from markupsafe import Markup, escape_silent as escape

class SilentMarkup(Markup):
    __slots__ = ()

    @classmethod
    def escape(cls, s):
        return cls(escape(s))

New-Style String Formatting

Starting with MarkupSafe 0.21 new style string formats from Python 2.6 and 3.x are now fully supported. Previously the escape behavior of those functions was spotty at best. The new implementations operates under the following algorithm:

  1. if an object has an __html_format__ method it is called as replacement for __format__ with the format specifier. It either has to return a string or markup object.
  2. if an object has an __html__ method it is called.
  3. otherwise the default format system of Python kicks in and the result is HTML escaped.

Here is how you can implement your own formatting:

class User(object):

    def __init__(self, id, username):
        self.id = id
        self.username = username

    def __html_format__(self, format_spec):
        if format_spec == 'link':
            return Markup('<a href="/user/{0}">{1}</a>').format(
                self.id,
                self.__html__(),
            )
        elif format_spec:
            raise ValueError('Invalid format spec')
        return self.__html__()

    def __html__(self):
        return Markup('<span class=user>{0}</span>').format(self.username)

And to format that user:

>>> user = User(1, 'foo')
>>> Markup('<p>User: {0:link}').format(user)
Markup(u'<p>User: <a href="/user/1"><span class=user>foo</span></a>')
Something went wrong with that request. Please try again.