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     """

    32     Returns the given HTML with ampersands, quotes and angle brackets encoded.

    33     """

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

    35 escape = allow_lazy(escape, unicode)

    36 

    37 def conditional_escape(html):

    38     """

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

    40     """

    41     if isinstance(html, SafeData):

    42         return html

    43     else:

    44         return escape(html)

    45 

    46 def linebreaks(value, autoescape=False):

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

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

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

    50     if autoescape:

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

    52     else:

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

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

    55 linebreaks = allow_lazy(linebreaks, unicode)

    56 

    57 def strip_tags(value):

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

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

    60 strip_tags = allow_lazy(strip_tags)

    61 

    62 def strip_spaces_between_tags(value):

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

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

    65 strip_spaces_between_tags = allow_lazy(strip_spaces_between_tags, unicode)

    66 

    67 def strip_entities(value):

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

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

    70 strip_entities = allow_lazy(strip_entities, unicode)

    71 

    72 def fix_ampersands(value):

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

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

    75 fix_ampersands = allow_lazy(fix_ampersands, unicode)

    76 

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

    78     """

    79     Converts any URLs in text into clickable links.

    80 

    81     Works on http://, https://, www. links and links ending in .org, .net or

    82     .com. Links can have trailing punctuation (periods, commas, close-parens)

    83     and leading punctuation (opening parens) and it'll still do the right

    84     thing.

    85 

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

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

    88 

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

    90     attribute.

    91 

    92     If autoescape is True, the link text and URLs will get autoescaped.

    93     """

    94     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

    95     safe_input = isinstance(text, SafeData)

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

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

    98     for i, word in enumerate(words):

    99         match = None

   100         if '.' in word or '@' in word or ':' in word:

   101             match = punctuation_re.match(word)

   102         if match:

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

   104             # Make URL we want to point to.

   105             url = None

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

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

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

   109                     middle and middle[0] in string.ascii_letters + string.digits and \

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

   111                 url = urlquote('http://%s' % middle, safe='/&=:;#?+*')

   112             elif '@' in middle and not ':' in middle and simple_email_re.match(middle):

   113                 url = 'mailto:%s' % middle

   114                 nofollow_attr = ''

   115             # Make link.

   116             if url:

   117                 trimmed = trim_url(middle)

   118                 if autoescape and not safe_input:

   119                     lead, trail = escape(lead), escape(trail)

   120                     url, trimmed = escape(url), escape(trimmed)

   121                 middle = '<a href="%s"%s>%s</a>' % (url, nofollow_attr, trimmed)

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

   123             else:

   124                 if safe_input:

   125                     words[i] = mark_safe(word)

   126                 elif autoescape:

   127                     words[i] = escape(word)

   128         elif safe_input:

   129             words[i] = mark_safe(word)

   130         elif autoescape:

   131             words[i] = escape(word)

   132     return u''.join(words)

   133 urlize = allow_lazy(urlize, unicode)

   134 

   135 def clean_html(text):

   136     """

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

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

   139         * Encode all ampersands correctly.

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

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

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

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

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

   145           bottom of the text.

   146     """

   147     from django.utils.text import normalize_newlines

   148     text = normalize_newlines(force_unicode(text))

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

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

   151     text = fix_ampersands(text)

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

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

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

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

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

   157     def replace_p_tags(match):

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

   159         for d in DOTS:

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

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

   162     text = hard_coded_bullets_re.sub(replace_p_tags, text)

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

   164     # of the text.

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

   166     return text

   167 clean_html = allow_lazy(clean_html, unicode)