html-sanitizer 1.1.3

Goals

HTML sanitizer goes further than e.g. bleach in that it not only ensures that content is safe and tags and attributes conform to a given whitelist, but also applies additional transforms to HTML fragments. A short list of goals follows:

• Clean up HTML using a very restricted set of allowed tags and attributes.

• Convert some tags (such as <span style="...">, <b> and <i>) into either <strong> or <em> (but never both).

• Normalize whitespace by removing repeated line breaks, empty paragraphs and other empty elements.

• Merge adjacent tags of the same type (such as several <strong> or <h3> directly after each other.

• Automatically remove redundant list markers inside <li> tags.

• Clean up some uglyness such as paragraphs inside paragraphs or list elements etc.

Usage

>>> from html_sanitizer import Sanitizer
>>> sanitizer = Sanitizer()  # default configuration
>>> sanitizer.sanitize('<span style="font-weight:bold">some text</span>')
'<strong>some text</strong>'

Settings

• span elements will always be removed from the tree, but only after inspecting their style tags (bold spans are converted into strong tags, italic spans into em tags)

• b and i tags will always be converted into strong and em (if strong and em are allowed at all)

The default settings are:

DEFAULT_SETTINGS = {
    'tags': {
        'a', 'h1', 'h2', 'h3', 'strong', 'em', 'p', 'ul', 'ol',
        'li', 'br', 'sub', 'sup', 'hr',
    },
    'attributes': {
        'a': ('href', 'name', 'target', 'title', 'id'),
    },
    'empty': {'hr', 'a', 'br'},
    'separate': {'a', 'p', 'li'},
    'add_nofollow': False,
    'autolink': False,
    'element_filters': [],
    'sanitize_href': html_sanitizer.sanitizer.sanitize_href,
}

The keys’ meaning is as follows:

• tags: A set() of allowed tags.

• attributes: A dict() mapping tags to their allowed attributes.

• empty: Tags which are allowed to be empty. By default, empty tags (containing no text or only whitespace) are dropped.

• separate: Tags which are not merged if they appear as siblings. By default, tags of the same type are merged.

• add_nofollow: Whether to add rel="nofollow" to all links.

• autolink: Enable lxml’s autolinker. May be either a boolean or a dictionary; a dictionary is passed as keyword arguments to autolink.

• element_filters: Additional filters that are called on all elements in the tree. The tree is processed in reverse depth-first order. Under certain circumstances elements are processed more than once (search the code for backlog.append)

• sanitize_href: A callable that gets anchor’s href value and returns a sanitized version. The default implementation checks whether links start with a few allowed prefixes, and if not, returns a single hash (#).

Settings can be specified partially when initializing a sanitizer instance, but are still checked for consistency (e.g. it’s not allowed to have tags in empty that are not in tags, that is, tags that are allowed to be empty but at the same time not allowed at all). An example for an even more restricted configuration might be:

>>> from html_sanitizer import Sanitizer
>>> sanitizer = Sanitizer({
...     'tags': ('h1', 'h2', 'p'),
...     'attributes': {},
...     'empty': set(),
...     'separate': set(),
... })

The rationale for such a restricted set of allowed tags (e.g. no images) is documented in the design decisions section of django-content-editor’s documentation.

Django

HTML sanitizer does not depend on Django, but ships with a module which makes configuring sanitizers using Django settings easier. Usage is as follows:

>>> from html_sanitizer.django import get_sanitizer
>>> sanitizer = get_sanitizer([name=...])

Different sanitizers can be configured. The default configuration is aptly named 'default'. Example settings follow:

HTML_SANITIZERS = {
    'default': {
    'tags': ...,
    ...
}

The 'default' configuration is special: If it isn’t explicitly defined, the default configuration above is used instead.