1 """HTML utilities suitable for global use."""

     2 

     3 import re

     4 import string

     5 

     6 from django.utils.safestring import SafeData, mark_safe

     7 from django.utils.encoding import force_unicode

     8 from django.utils.functional import allow_lazy

     9 from django.utils.http import urlquote

    10 

    11 # Configuration for urlize() function.

    12 LEADING_PUNCTUATION  = ['(', '<', '&lt;']

    13 TRAILING_PUNCTUATION = ['.', ',', ')', '>', '\n', '>']

    14 

    15 # List of possible strings used for bullets in bulleted lists.

    16 DOTS = ['·', '*', '\xe2\x80\xa2', '•', '•', '•']

    17 

    18 unencoded_ampersands_re = re.compile(r'&(?!(\w+|#\d+);)')

    19 word_split_re = re.compile(r'(\s+)')

    20 punctuation_re = re.compile('^(?P<lead>(?:%s)*)(?P<middle>.*?)(?P<trail>(?:%s)*)$' % \

    21     ('|'.join([re.escape(x) for x in LEADING_PUNCTUATION]),

    22     '|'.join([re.escape(x) for x in TRAILING_PUNCTUATION])))

    23 simple_email_re = re.compile(r'^\S+@[a-zA-Z0-9._-]+\.[a-zA-Z0-9._-]+$')

    24 link_target_attribute_re = re.compile(r'(<a [^>]*?)target=[^\s>]+')

    25 html_gunk_re = re.compile(r'(?:<br clear="all">|<i><\/i>|<b><\/b>|<em><\/em>|<strong><\/strong>|<\/?smallcaps>|<\/?uppercase>)', re.IGNORECASE)

    26 hard_coded_bullets_re = re.compile(r'((?:<p>(?:%s).*?[a-zA-Z].*?</p>\s*)+)' % '|'.join([re.escape(x) for x in DOTS]), re.DOTALL)

    27 trailing_empty_content_re = re.compile(r'(?:<p>(?:&nbsp;|\s|<br \/>)*?</p>\s*)+\Z')

    28 del x # Temporary variable

    29 

    30 def escape(html):

    31     """Returns the given HTML with ampersands, quotes and carets encoded."""

    32     return mark_safe(force_unicode(html).replace('&', '&amp;').replace('<', '&lt;').replace('>', '&gt;').replace('"', '&quot;').replace("'", '&#39;'))

    33 escape = allow_lazy(escape, unicode)

    34 

    35 def conditional_escape(html):

    36     """

    37     Similar to escape(), except that it doesn't operate on pre-escaped strings.

    38     """

    39     if isinstance(html, SafeData):

    40         return html

    41     else:

    42         return escape(html)

    43 

    44 def linebreaks(value, autoescape=False):

    45     """Converts newlines into <p> and <br />s."""

    46     value = re.sub(r'\r\n|\r|\n', '\n', force_unicode(value)) # normalize newlines

    47     paras = re.split('\n{2,}', value)

    48     if autoescape:

    49         paras = [u'<p>%s</p>' % escape(p.strip()).replace('\n', '<br />') for p in paras]

    50     else:

    51         paras = [u'<p>%s</p>' % p.strip().replace('\n', '<br />') for p in paras]

    52     return u'\n\n'.join(paras)

    53 linebreaks = allow_lazy(linebreaks, unicode)

    54 

    55 def strip_tags(value):

    56     """Returns the given HTML with all tags stripped."""

    57     return re.sub(r'<[^>]*?>', '', force_unicode(value))

    58 strip_tags = allow_lazy(strip_tags)

    59 

    60 def strip_spaces_between_tags(value):

    61     """Returns the given HTML with spaces between tags removed."""

    62     return re.sub(r'>\s+<', '><', force_unicode(value))

    63 strip_spaces_between_tags = allow_lazy(strip_spaces_between_tags, unicode)

    64 

    65 def strip_entities(value):

    66     """Returns the given HTML with all entities (&something;) stripped."""

    67     return re.sub(r'&(?:\w+|#\d+);', '', force_unicode(value))

    68 strip_entities = allow_lazy(strip_entities, unicode)

    69 

    70 def fix_ampersands(value):

    71     """Returns the given HTML with all unencoded ampersands encoded correctly."""

    72     return unencoded_ampersands_re.sub('&', force_unicode(value))

    73 fix_ampersands = allow_lazy(fix_ampersands, unicode)

    74 

    75 def urlize(text, trim_url_limit=None, nofollow=False, autoescape=False):

    76     """

    77     Converts any URLs in text into clickable links.

    78 

    79     Works on http://, https://, and www. links.  Links can have trailing

    80     punctuation (periods, commas, close-parens) and leading punctuation

    81     (opening parens) and it'll still do the right thing.

    82 

    83     If trim_url_limit is not None, the URLs in link text longer than this limit

    84     will truncated to trim_url_limit-3 characters and appended with an elipsis.

    85 

    86     If nofollow is True, the URLs in link text will get a rel="nofollow"

    87     attribute.

    88     """

    89     if autoescape:

    90         trim_url = lambda x, limit=trim_url_limit: conditional_escape(limit is not None and (len(x) > limit and ('%s...' % x[:max(0, limit - 3)])) or x)

    91     else:

    92         trim_url = lambda x, limit=trim_url_limit: limit is not None and (len(x) > limit and ('%s...' % x[:max(0, limit - 3)])) or x

    93     safe_input = isinstance(text, SafeData)

    94     words = word_split_re.split(force_unicode(text))

    95     nofollow_attr = nofollow and ' rel="nofollow"' or ''

    96     for i, word in enumerate(words):

    97         match = punctuation_re.match(word)

    98         if match:

    99             lead, middle, trail = match.groups()

   100             if safe_input:

   101                 middle = mark_safe(middle)

   102             if middle.startswith('www.') or ('@' not in middle and not middle.startswith('http://') and \

   103                     len(middle) > 0 and middle[0] in string.ascii_letters + string.digits and \

   104                     (middle.endswith('.org') or middle.endswith('.net') or middle.endswith('.com'))):

   105                 middle = 'http://%s' % middle

   106             if middle.startswith('http://') or middle.startswith('https://'):

   107                 url = urlquote(middle, safe='/&=:;#?+*')

   108                 if autoescape and not safe_input:

   109                     url = escape(url)

   110                 trimmed_url = trim_url(middle)

   111                 middle = '<a href="%s"%s>%s</a>' % (url, nofollow_attr,

   112                         trimmed_url)

   113             elif '@' in middle and not middle.startswith('www.') and \

   114                       not ':' in middle and simple_email_re.match(middle):

   115                 if autoescape:

   116                     middle = conditional_escape(middle)

   117                 middle = '<a href="mailto:%s">%s</a>' % (middle, middle)

   118             if lead + middle + trail != word:

   119                 if autoescape and not safe_input:

   120                     lead, trail = escape(lead), escape(trail)

   121                 words[i] = mark_safe('%s%s%s' % (lead, middle, trail))

   122             elif autoescape and not safe_input:

   123                 words[i] = escape(word)

   124         elif safe_input:

   125             words[i] = mark_safe(word)

   126         elif autoescape:

   127             words[i] = escape(word)

   128     return u''.join(words)

   129 urlize = allow_lazy(urlize, unicode)

   130 

   131 def clean_html(text):

   132     """

   133     Clean the given HTML.  Specifically, do the following:

   134         * Convert <b> and <i> to <strong> and <em>.

   135         * Encode all ampersands correctly.

   136         * Remove all "target" attributes from <a> tags.

   137         * Remove extraneous HTML, such as presentational tags that open and

   138           immediately close and <br clear="all">.

   139         * Convert hard-coded bullets into HTML unordered lists.

   140         * Remove stuff like "<p>&nbsp;&nbsp;</p>", but only if it's at the

   141           bottom of the text.

   142     """

   143     from django.utils.text import normalize_newlines

   144     text = normalize_newlines(force_unicode(text))

   145     text = re.sub(r'<(/?)\s*b\s*>', '<\\1strong>', text)

   146     text = re.sub(r'<(/?)\s*i\s*>', '<\\1em>', text)

   147     text = fix_ampersands(text)

   148     # Remove all target="" attributes from <a> tags.

   149     text = link_target_attribute_re.sub('\\1', text)

   150     # Trim stupid HTML such as <br clear="all">.

   151     text = html_gunk_re.sub('', text)

   152     # Convert hard-coded bullets into HTML unordered lists.

   153     def replace_p_tags(match):

   154         s = match.group().replace('</p>', '</li>')

   155         for d in DOTS:

   156             s = s.replace('<p>%s' % d, '<li>')

   157         return u'<ul>\n%s\n</ul>' % s

   158     text = hard_coded_bullets_re.sub(replace_p_tags, text)

   159     # Remove stuff like "<p>&nbsp;&nbsp;</p>", but only if it's at the bottom

   160     # of the text.

   161     text = trailing_empty_content_re.sub('', text)

   162     return text

   163 clean_html = allow_lazy(clean_html, unicode)