--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/gpp/legacy/html2md.py	Fri Dec 24 05:28:58 2010 +0000
@@ -0,0 +1,279 @@
+"""
+This module contains a class derived from Python's HTMLParser to convert HTML to
+Markdown. Currently this class only supports those HTML tags that have counter-
+parts in BBCode used by stock phpBB 2.x.
+
+In other words, this class was created to help convert data from a phpBB
+forum to Markdown syntax and its scope is currently limited to that task.
+
+"""
+from HTMLParser import HTMLParser
+import htmlentitydefs
+
+
+# Let's call Markdown markup entities "elements" to avoid confusion
+# with HTML tags.
+
+class ElementBase(object):
+    """
+    Base class for all Markdown elements.
+
+    """
+    def __init__(self, attrs=None):
+        self.data = u''
+        self.attrs = dict(attrs) if attrs else {}
+
+    def add_data(self, data):
+        self.data += data
+
+    def markdown(self):
+        return self.data
+
+
+class TextElement(ElementBase):
+    """
+    TextElements represent text fragments not inside HTML tags.
+    """
+    pass
+
+
+class EmphasisElement(ElementBase):
+    """
+    An EmphasisElement is a Markdown element used to indicate emphasis and is
+    represented by placing characters around text. E.g. _em_, **bold**
+
+    """
+    def __init__(self, tag, attrs):
+        super(EmphasisElement, self).__init__(attrs)
+        self.tag = tag
+
+    def markdown(self):
+        return u'%s%s%s' % (self.tag, self.data, self.tag)
+
+
+def create_emphasis(tag):
+    """
+    Returns a function that creates an EmphasisElement using the supplied
+    tag.
+
+    """
+    def inner(attrs):
+        return EmphasisElement(tag, attrs)
+    return inner
+
+
+class HtmlElement(ElementBase):
+    """
+    Markdown also accepts HTML markup. This element represents a HTML tag that
+    maps to itself in Markdown.
+
+    """
+    def __init__(self, tag, attrs):
+        super(HtmlElement, self).__init__(attrs)
+        self.tag = tag
+
+    def markdown(self):
+        return u'<%s>%s</%s>' % (self.tag, self.data, self.tag)
+
+
+def create_html(tag):
+    """
+    Returns a function that creates a HtmlElement using the supplied tag.
+
+    """
+    def inner(attrs):
+        return HtmlElement(tag, attrs)
+    return inner
+
+
+class QuoteElement(ElementBase):
+    """
+    Class to represent a blockquote in Markdown.
+
+    """
+    def markdown(self):
+        return u'> %s\n\n' % self.data.replace('\n', '\n> ')
+
+
+class BreakElement(ElementBase):
+    """
+    Class to represent a linebreak in Markdown.
+
+    """
+    def markdown(self):
+        return u'  \n'
+
+
+class DivElement(ElementBase):
+    """
+    This class maps a HTML <div> into a block of text surrounded by newlines.
+
+    """
+    def markdown(self):
+        return u'\n%s\n' % self.data
+
+
+class LinkElement(ElementBase):
+    """
+    This class maps HTML <a> tags into Markdown links.
+    If no data is present, the actual href is used for the link text.
+
+    """
+    def markdown(self):
+        url = self.attrs['href']
+        text = self.data if self.data else url
+        return u'[%s](%s)' % (text, url)
+
+
+class ImageElement(ElementBase):
+    """
+    This class maps HTML <img> tags into Markdown.
+    This element assumes no alt text is present, and simply uses the word
+    'image' for the alt text.
+
+    """
+    def markdown(self):
+        url = self.attrs['src']
+        return u'![image](%s)' % url
+
+
+class CodeElement(ElementBase):
+    """
+    This class is used to create code blocks in Markdown.
+
+    """
+    def markdown(self):
+        return u'    %s\n' % self.data.replace('\n', '\n    ')
+
+
+# List (ordered & unordered) support:
+
+class ListElement(ElementBase):
+    """
+    This class creates Markdown for unordered lists. The bullet() method can be
+    overridden to create ordered lists.
+
+    """
+    def __init__(self, attrs=None):
+        super(ListElement, self).__init__(attrs)
+        self.items = []
+        self.list_nesting = 1
+
+    def add_data(self, data):
+        self.items.append(data)
+
+    def bullet(self):
+        return u'*'
+
+    def markdown(self):
+        bullet_str = self.bullet()
+        indent = u' ' * (4 * (self.list_nesting - 1))
+        s = u''
+        for item in self.items:
+            s += u'\n%s%s %s' % (indent, bullet_str, item)
+        return s
+
+
+class OrderedListElement(ListElement):
+    """
+    This class creates Markdown for ordered lists.
+
+    """
+    def bullet(self):
+        return '1.'
+
+
+class ItemElement(ElementBase):
+    """
+    This element is used to represent ordered & unordered list items.
+
+    """
+    pass
+
+###############################################################################
+###############################################################################
+
+class MarkdownWriter(HTMLParser):
+    """
+    This class is an HTMLParser that converts a subset of HTML to Markdown.
+
+    """
+
+    elem_factories = {
+        'a': LinkElement,
+        'blockquote': QuoteElement,
+        'br': BreakElement,
+        'div': DivElement,
+        'em': create_emphasis('_'),
+        'img': ImageElement,
+        'li': ItemElement,
+        'ol': OrderedListElement,
+        'pre': CodeElement,
+        's': create_html('strike'),
+        'strong': create_emphasis('**'),
+        'u': create_html('u'),
+        'ul': ListElement,
+    }
+
+    def __init__(self):
+        HTMLParser.__init__(self)
+        self.reset()
+
+    def handle_starttag(self, tag, attrs):
+        if tag in self.elem_factories:
+            factory = self.elem_factories[tag]
+            element = factory(attrs)
+        else:
+            element = TextElement()
+
+        self._push_elem(element)
+
+    def handle_endtag(self, tag):
+        self._pop_elem()
+
+    def handle_data(self, data):
+        if len(self.elem_stack) == 0:
+            self._push_elem(TextElement())
+        self._add_data(data)
+
+    def handle_entityref(self, name):
+        try:
+            text = unichr(htmlentitydefs.name2codepoint[name])
+        except KeyError:
+            text = name
+        self.handle_data(text)
+
+    def handle_charref(self, name):
+        self.handle_data(unichr(int(name)))
+
+    def reset(self):
+        HTMLParser.reset(self)
+        self.elem_stack = []
+        self.elements = []
+        self.list_nesting = 0
+
+    def _push_elem(self, tag):
+        if len(self.elem_stack) and isinstance(self.elem_stack[-1], TextElement):
+            self._pop_elem()
+        if isinstance(tag, ListElement):
+            self.list_nesting += 1
+            tag.list_nesting = self.list_nesting
+        self.elem_stack.append(tag)
+
+    def _pop_elem(self):
+        element = self.elem_stack.pop()
+        if isinstance(element, ListElement):
+            self.list_nesting -= 1
+        if len(self.elem_stack):
+            self.elem_stack[-1].add_data(element.markdown())
+        else:
+            self.elements.append(element)
+
+    def _add_data(self, data):
+        self.elem_stack[-1].add_data(data)
+
+    def markdown(self):
+        while len(self.elem_stack):
+            self._pop_elem()
+        text_list = [e.markdown() for e in self.elements]
+        return u''.join(text_list)