"""
POST-PROCESSORS
=============================================================================

Markdown also allows post-processors, which are similar to preprocessors in
that they need to implement a "run" method. However, they are run after core
processing.

"""

import re
import util
import odict

def build_postprocessors(md_instance, **kwargs):
    """ Build the default postprocessors for Markdown. """
    postprocessors = odict.OrderedDict()
    postprocessors["raw_html"] = RawHtmlPostprocessor(md_instance)
    postprocessors["amp_substitute"] = AndSubstitutePostprocessor()
    postprocessors["unescape"] = UnescapePostprocessor()
    return postprocessors


class Postprocessor(util.Processor):
    """
    Postprocessors are run after the ElementTree it converted back into text.

    Each Postprocessor implements a "run" method that takes a pointer to a
    text string, modifies it as necessary and returns a text string.

    Postprocessors must extend markdown.Postprocessor.

    """

    def run(self, text):
        """
        Subclasses of Postprocessor should implement a `run` method, which
        takes the html document as a single text string and returns a
        (possibly modified) string.

        """
        pass


class RawHtmlPostprocessor(Postprocessor):
    """ Restore raw html to the document. """

    def run(self, text):
        """ Iterate over html stash and restore "safe" html. """
        for i in range(self.markdown.htmlStash.html_counter):
            html, safe  = self.markdown.htmlStash.rawHtmlBlocks[i]
            if self.markdown.safeMode and not safe:
                if str(self.markdown.safeMode).lower() == 'escape':
                    html = self.escape(html)
                elif str(self.markdown.safeMode).lower() == 'remove':
                    html = ''
                else:
                    html = self.markdown.html_replacement_text
            if self.isblocklevel(html) and (safe or not self.markdown.safeMode):
                text = text.replace("<p>%s</p>" % 
                            (self.markdown.htmlStash.get_placeholder(i)),
                            html + "\n")
            html = self.unescape(html)
            text =  text.replace(self.markdown.htmlStash.get_placeholder(i), 
                                 html)
        return text

    def unescape(self, html):
        """ Unescape any markdown escaped text within inline html. """
        for k, v in self.markdown.treeprocessors['inline'].stashed_nodes.items():
            ph = util.INLINE_PLACEHOLDER % k
            try:
                html = html.replace(ph, '%s' % util.etree.tostring(v))
            except:
                html = html.replace(ph, '\%s' % v)
        return html

    def escape(self, html):
        """ Basic html escaping """
        html = html.replace('&', '&amp;')
        html = html.replace('<', '&lt;')
        html = html.replace('>', '&gt;')
        return html.replace('"', '&quot;')


    def isblocklevel(self, html):
        m = re.match(r'^\<\/?([^ ]+)', html)
        if m:
            if m.group(1)[0] in ('!', '?', '@', '%'):
                # Comment, php etc...
                return True
            return util.isBlockLevel(m.group(1))
        return False


class AndSubstitutePostprocessor(Postprocessor):
    """ Restore valid entities """

    def run(self, text):
        text =  text.replace(util.AMP_SUBSTITUTE, "&")
        return text


class UnescapePostprocessor(Postprocessor):
    """ Restore escaped chars """

    RE = re.compile('%s(\d+)%s' % (util.STX, util.ETX))

    def unescape(self, m):
        return unichr(int(m.group(1)))

    def run(self, text):
        return self.RE.sub(self.unescape, text)