Function normalize_html

test/normalize.py:131–194

r""" Return normalized form of HTML which ignores insignificant output differences: Multiple inner whitespaces are collapsed to a single space (except in pre tags): >>> normalize_html(" a \t b ") ' a b ' >>> normalize_html(" a \t\nb ")

(html)

            'h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'video', 'script', 'style'])

def normalize_html(html):
    r"""
    Return normalized form of HTML which ignores insignificant output
    differences:

    Multiple inner whitespaces are collapsed to a single space (except
    in pre tags):

        >>> normalize_html("<p>a  \t b</p>")
        '<p>a b</p>'

        >>> normalize_html("<p>a  \t\nb</p>")
        '<p>a b</p>'

    * Whitespace surrounding block-level tags is removed.

        >>> normalize_html("<p>a  b</p>")
        '<p>a b</p>'

        >>> normalize_html(" <p>a  b</p>")
        '<p>a b</p>'

        >>> normalize_html("<p>a  b</p> ")
        '<p>a b</p>'

        >>> normalize_html("\n\t<p>\n\t\ta  b\t\t</p>\n\t")
        '<p>a b</p>'

        >>> normalize_html("<i>a  b</i> ")
        '<i>a b</i> '

    * Self-closing tags are converted to open tags.

        >>> normalize_html("<br />")
        '<br>'

    * Attributes are sorted and lowercased.

        >>> normalize_html('<a title="bar" HREF="foo">x</a>')
        '<a href="foo" title="bar">x</a>'

    * References are converted to unicode, except that '<', '>', '&', and
      '"' are rendered using entities.

        >>> normalize_html("&forall;&amp;&gt;&lt;&quot;")
        '\u2200&amp;&gt;&lt;&quot;'

    """
    html_chunk_re = re.compile(r"(\<!\[CDATA\[.*?\]\]\>|\<[^>]*\>|[^<]+)")
    try:
        parser = MyHTMLParser()
        # We work around HTMLParser's limitations parsing CDATA
        # by breaking the input into chunks and passing CDATA chunks
        # through verbatim.
        for chunk in re.finditer(html_chunk_re, html):
            if chunk.group(0)[:8] == "<![CDATA":
                parser.output += chunk.group(0)
            else: