0
@@ -0,0 +1,251 @@
0
+# from django.conf import settings
0
+import re
0
+from django.conf import settings
0
+from django import template
0
+from django.template.defaultfilters import stringfilter
0
+
0
+register = template.Library()
0
+
0
+def amp(text):
0
+    """Wraps apersands in HTML with ``<span class="amp">`` so they can be
0
+    styled with CSS. Apersands are also normalized to ``&amp;``. Requires 
0
+    ampersands to have whitespace or an ``&nbsp;`` on both sides.
0
+    
0
+    >>> amp('One & two')
0
+    'One <span class="amp">&amp;</span> two'
0
+    >>> amp('One &amp; two')
0
+    'One <span class="amp">&amp;</span> two'
0
+    >>> amp('One &#38; two')
0
+    'One <span class="amp">&amp;</span> two'
0
+
0
+    >>> amp('One&nbsp;&amp;&nbsp;two')
0
+    'One&nbsp;<span class="amp">&amp;</span>&nbsp;two'
0
+
0
+    It won't mess up & that are already wrapped, in entities or URLs
0
+
0
+    >>> amp('One <span class="amp">&amp;</span> two')
0
+    'One <span class="amp">&amp;</span> two'
0
+    >>> amp('&ldquo;this&rdquo; & <a href="/?that&amp;test">that</a>')
0
+    '&ldquo;this&rdquo; <span class="amp">&amp;</span> <a href="/?that&amp;test">that</a>'
0
+
0
+    It should ignore standalone amps that are in attributes
0
+    >>> amp('<link href="xyz.html" title="One & Two">xyz</link>')
0
+    '<link href="xyz.html" title="One & Two">xyz</link>'
0
+    """
0
+    # tag_pattern from http://haacked.com/archive/2004/10/25/usingregularexpressionstomatchhtml.aspx
0
+    # it kinda sucks but it fixes the standalone amps in attributes bug
0
+    tag_pattern = '</?\w+((\s+\w+(\s*=\s*(?:".*?"|\'.*?\'|[^\'">\s]+))?)+\s*|\s*)/?>'
0
+    amp_finder = re.compile(r"(\s|&nbsp;)(&|&amp;|&\#38;)(\s|&nbsp;)")
0
+    intra_tag_finder = re.compile(r'(?P<prefix>(%s)?)(?P<text>([^<]*))(?P<suffix>(%s)?)' % (tag_pattern, tag_pattern))
0
+    def _amp_process(groups):
0
+        prefix = groups.group('prefix') or ''
0
+        text = amp_finder.sub(r"""\1<span class="amp">&amp;</span>\3""", groups.group('text'))
0
+        suffix = groups.group('suffix') or ''
0
+        return prefix + text + suffix
0
+    return intra_tag_finder.sub(_amp_process, text)
0
+# amp = stringfilter(amp)
0
+
0
+def caps(text):
0
+    """Wraps multiple capital letters in ``<span class="caps">`` 
0
+    so they can be styled with CSS. 
0
+    
0
+    >>> caps("A message from KU")
0
+    'A message from <span class="caps">KU</span>'
0
+    
0
+    Uses the smartypants tokenizer to not screw with HTML or with tags it shouldn't.
0
+    
0
+    >>> caps("<PRE>CAPS</pre> more CAPS")
0
+    '<PRE>CAPS</pre> more <span class="caps">CAPS</span>'
0
+
0
+    >>> caps("A message from 2KU2 with digits")
0
+    'A message from <span class="caps">2KU2</span> with digits'
0
+        
0
+    >>> caps("Dotted caps followed by spaces should never include them in the wrap D.O.T.   like so.")
0
+    'Dotted caps followed by spaces should never include them in the wrap <span class="caps">D.O.T.</span>  like so.'
0
+
0
+    >>> caps("<i>D.O.T.</i>HE34T<b>RFID</b>")
0
+    '<i><span class="caps">D.O.T.</span></i><span class="caps">HE34T</span><b><span class="caps">RFID</span></b>'
0
+    """
0
+    try:
0
+        import smartypants
0
+    except ImportError:
0
+        if settings.DEBUG:
0
+            raise template.TemplateSyntaxError, "Error in {% caps %} filter: The Python SmartyPants library isn't installed."
0
+        return text
0
+        
0
+    tokens = smartypants._tokenize(text)
0
+    result = []
0
+    in_skipped_tag = False    
0
+    
0
+    cap_finder = re.compile(r"""(
0
+                            (\b[A-Z\d]*        # Group 2: Any amount of caps and digits
0
+                            [A-Z]\d*[A-Z]      # A cap string much at least include two caps (but they can have digits between them)
0
+                            [A-Z\d]*\b)        # Any amount of caps and digits
0
+                            | (\b[A-Z]+\.\s?   # OR: Group 3: Some caps, followed by a '.' and an optional space
0
+                            (?:[A-Z]+\.\s?)+)  # Followed by the same thing at least once more
0
+                            (?:\s|\b|$))
0
+                            """, re.VERBOSE)
0
+
0
+    def _cap_wrapper(matchobj):
0
+        """This is necessary to keep dotted cap strings to pick up extra spaces"""
0
+        if matchobj.group(2):
0
+            return """<span class="caps">%s</span>""" % matchobj.group(2)
0
+        else:
0
+            if matchobj.group(3)[-1] == " ":
0
+                caps = matchobj.group(3)[:-1]
0
+                tail = ' '
0
+            else:
0
+                caps = matchobj.group(3)
0
+                tail = ''
0
+            return """<span class="caps">%s</span>%s""" % (caps, tail)
0
+
0
+    tags_to_skip_regex = re.compile("<(/)?(?:pre|code|kbd|script|math)[^>]*>", re.IGNORECASE)
0
+    
0
+    
0
+    for token in tokens:
0
+        if token[0] == "tag":
0
+            # Don't mess with tags.
0
+            result.append(token[1])
0
+            close_match = tags_to_skip_regex.match(token[1])
0
+            if close_match and close_match.group(1) == None:
0
+                in_skipped_tag = True
0
+            else:
0
+                in_skipped_tag = False
0
+        else:
0
+            if in_skipped_tag:
0
+                result.append(token[1])
0
+            else:
0
+                result.append(cap_finder.sub(_cap_wrapper, token[1]))
0
+            
0
+    return "".join(result)
0
+# caps = stringfilter(caps)
0
+
0
+def initial_quotes(text):
0
+    """Wraps initial quotes in ``class="dquo"`` for double quotes or  
0
+    ``class="quo"`` for single quotes. Works in these block tags ``(h1-h6, p, li, dt, dd)``
0
+    and also accounts for potential opening inline elements ``a, em, strong, span, b, i``
0
+    
0
+    >>> initial_quotes('"With primes"')
0
+    '<span class="dquo">"</span>With primes"'
0
+    >>> initial_quotes("'With single primes'")
0
+    '<span class="quo">\\'</span>With single primes\\''
0
+    
0
+    >>> initial_quotes('<a href="#">"With primes and a link"</a>')
0
+    '<a href="#"><span class="dquo">"</span>With primes and a link"</a>'
0
+    
0
+    >>> initial_quotes('&#8220;With smartypanted quotes&#8221;')
0
+    '<span class="dquo">&#8220;</span>With smartypanted quotes&#8221;'
0
+    """
0
+    quote_finder = re.compile(r"""((<(p|h[1-6]|li|dt|dd)[^>]*>|^)              # start with an opening p, h1-6, li, dd, dt or the start of the string
0
+                                  \s*                                          # optional white space! 
0
+                                  (<(a|em|span|strong|i|b)[^>]*>\s*)*)         # optional opening inline tags, with more optional white space for each.
0
+                                  (("|&ldquo;|&\#8220;)|('|&lsquo;|&\#8216;))  # Find me a quote! (only need to find the left quotes and the primes)
0
+                                                                               # double quotes are in group 7, singles in group 8 
0
+                                  """, re.VERBOSE)
0
+    def _quote_wrapper(matchobj):
0
+        if matchobj.group(7): 
0
+            classname = "dquo"
0
+            quote = matchobj.group(7)
0
+        else:
0
+            classname = "quo"
0
+            quote = matchobj.group(8)
0
+        return """%s<span class="%s">%s</span>""" % (matchobj.group(1), classname, quote) 
0
+        
0
+    return quote_finder.sub(_quote_wrapper, text)
0
+# initial_quotes = stringfilter(initial_quotes)
0
+
0
+def smartypants(text):
0
+    """Applies smarty pants to curl quotes.
0
+    
0
+    >>> smartypants('The "Green" man')
0
+    'The &#8220;Green&#8221; man'
0
+    """
0
+    try:
0
+        import smartypants
0
+    except ImportError:
0
+        if settings.DEBUG:
0
+            raise template.TemplateSyntaxError, "Error in {% smartypants %} filter: The Python smartypants library isn't installed."
0
+        return text
0
+    else:
0
+        return smartypants.smartyPants(text)
0
+# smartypants = stringfilter(smartypants)
0
+
0
+def typogrify(text):
0
+    """The super typography filter
0
+    
0
+    Applies the following filters: widont, smartypants, caps, amp, initial_quotes
0
+    
0
+    >>> typogrify('<h2>"Jayhawks" & KU fans act extremely obnoxiously</h2>')
0
+    '<h2><span class="dquo">&#8220;</span>Jayhawks&#8221; <span class="amp">&amp;</span> <span class="caps">KU</span> fans act extremely&nbsp;obnoxiously</h2>'
0
+    """
0
+    text = amp(text)
0
+    text = widont(text)
0
+    text = smartypants(text)
0
+    text = caps(text)
0
+    text = initial_quotes(text)
0
+    return text
0
+# typogrify = stringfilter(typogrify)
0
+
0
+def widont(text):
0
+    """Replaces the space between the last two words in a string with ``&nbsp;``
0
+    Works in these block tags ``(h1-h6, p, li, dd, dt)`` and also accounts for 
0
+    potential closing inline elements ``a, em, strong, span, b, i``
0
+    
0
+    >>> widont('A very simple test')
0
+    'A very simple&nbsp;test'
0
+
0
+    Single word items shouldn't be changed
0
+    >>> widont('Test')
0
+    'Test'
0
+    >>> widont(' Test')
0
+    ' Test'
0
+    >>> widont('<ul><li>Test</p></li><ul>')
0
+    '<ul><li>Test</p></li><ul>'
0
+    >>> widont('<ul><li> Test</p></li><ul>')
0
+    '<ul><li> Test</p></li><ul>'
0
+    
0
+    >>> widont('<p>In a couple of paragraphs</p><p>paragraph two</p>')
0
+    '<p>In a couple of&nbsp;paragraphs</p><p>paragraph&nbsp;two</p>'
0
+    
0
+    >>> widont('<h1><a href="#">In a link inside a heading</i> </a></h1>')
0
+    '<h1><a href="#">In a link inside a&nbsp;heading</i> </a></h1>'
0
+    
0
+    >>> widont('<h1><a href="#">In a link</a> followed by other text</h1>')
0
+    '<h1><a href="#">In a link</a> followed by other&nbsp;text</h1>'
0
+
0
+    Empty HTMLs shouldn't error
0
+    >>> widont('<h1><a href="#"></a></h1>') 
0
+    '<h1><a href="#"></a></h1>'
0
+    
0
+    >>> widont('<div>Divs get no love!</div>')
0
+    '<div>Divs get no love!</div>'
0
+    
0
+    >>> widont('<pre>Neither do PREs</pre>')
0
+    '<pre>Neither do PREs</pre>'
0
+    
0
+    >>> widont('<div><p>But divs with paragraphs do!</p></div>')
0
+    '<div><p>But divs with paragraphs&nbsp;do!</p></div>'
0
+    """
0
+    widont_finder = re.compile(r"""((?:</?(?:a|em|span|strong|i|b)[^>]*>)|[^<>\s]) # must be proceeded by an approved inline opening or closing tag or a nontag/nonspace
0
+                                   \s+                                # the space to replace
0
+                                   ([^<>\s]+                            # must be flollowed by non-tag non-space characters
0
+                                   \s*                                  # optional white space! 
0
+                                   (</(a|em|span|strong|i|b)>\s*)* # optional closing inline tags with optional white space after each
0
+                                   ((</(p|h[1-6]|li|dt|dd)>)|$))                 # end with a closing p, h1-6, li or the end of the string
0
+                                   """, re.VERBOSE)
0
+    return widont_finder.sub(r'\1&nbsp;\2', text)
0
+# widont = stringfilter(widont)
0
+
0
+register.filter('amp', amp)
0
+register.filter('caps', caps)
0
+register.filter('initial_quotes', initial_quotes)
0
+register.filter('smartypants', smartypants)
0
+register.filter('typogrify', typogrify)
0
+register.filter('widont', widont)
0
+
0
+def _test():
0
+    import doctest
0
+    doctest.testmod()
0
+
0
+if __name__ == "__main__":
0
+    _test()