python 将html转换为markdown，pythonmarkdown,markdown格式要比

python 将html转换为markdown，pythonmarkdown,markdown格式要比

markdown格式要比html格式清爽的多，所以在某些场合需要将html转换为markdown，下面是一段转换的代码片段，来自git，有修改：

#!/usr/bin/env python#encoding=utf8from HTMLParser import HTMLParserimport reimport osimport sysclass Html2MarkdownParser(HTMLParser):    def __init__(self):        self._markdown = ''        self._tag_stack = []        self._tag_attr_data = {}        self._handled_tag_body_data = ''        self._convertible_tags = ['a',                                  'b', 'blockquote',                                  'em',                                  'h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'hr',                                  'ol',                                  'p', 'pre',                                  'strong',                                  'ul']        # FIXME: special characters        HTMLParser.__init__(self)    def _append_to_markdown(self, new_markdown):        if len(self._markdown) > 1:            if re.match('\s', self._markdown[-1:]):                self._markdown += new_markdown            else:                self._markdown += ' ' + new_markdown        else:            self._markdown += new_markdown    # <a />    def handle_start_a(self, attrs):        self._tag_attr_data = dict(attrs)    def handle_end_a(self):        a_tag = ''        a_tag += '[' + self._handled_tag_body_data + ']'        a_tag += '(' + self._tag_attr_data.get('href')        title = self._tag_attr_data.get('title')        if title:            a_tag += ' "' + title + '") '        else:            a_tag += ') '        self._append_to_markdown(a_tag)    # <b />    def handle_end_b(self):        self._handled_tag_body_data = self._handled_tag_body_data.replace(os.linesep, ' ')        self._append_to_markdown('*' + self._handled_tag_body_data + '*')    # <blockquote />    def handle_end_blockquote(self):        blockquote_body = self._handled_tag_body_data.split(os.linesep)        for blockquote_line in blockquote_body:            blockquote_line = blockquote_line.strip()            self._append_to_markdown('> ' + blockquote_line + os.linesep)    # <em />    def handle_end_em(self):        self._handled_tag_body_data = self._handled_tag_body_data.replace(os.linesep, ' ')        self._append_to_markdown('*' + self._handled_tag_body_data + '*')    # <h1 />    def handle_end_h1(self):        self._handled_tag_body_data = self._handled_tag_body_data.replace(os.linesep, ' ')        self._append_to_markdown('# ' + self._handled_tag_body_data + ' #' + os.linesep)    # <h2 />    def handle_end_h2(self):        self._handled_tag_body_data = self._handled_tag_body_data.replace(os.linesep, ' ')        self._append_to_markdown('## ' + self._handled_tag_body_data + ' ##' + os.linesep)    # <h3 />    def handle_end_h3(self):        self._handled_tag_body_data = self._handled_tag_body_data.replace(os.linesep, ' ')        self._append_to_markdown('### ' + self._handled_tag_body_data + ' ###' + os.linesep)    # <h4 />    def handle_end_h4(self):        self._handled_tag_body_data = self._handled_tag_body_data.replace(os.linesep, ' ')        self._append_to_markdown('#### ' + self._handled_tag_body_data + ' ####' + os.linesep)    # <h5 />    def handle_end_h5(self):        self._handled_tag_body_data = self._handled_tag_body_data.replace(os.linesep, ' ')        self._append_to_markdown('##### ' + self._handled_tag_body_data + ' #####' + os.linesep)    # <h6 />    def handle_end_h6(self):        self._handled_tag_body_data = self._handled_tag_body_data.replace(os.linesep, ' ')        self._append_to_markdown('###### ' + self._handled_tag_body_data + ' ######' + os.linesep)    # <hr />    def handle_start_hr(self, attrs):        self._append_to_markdown('* * *' + os.linesep)    # <li />    def handle_end_li(self):        if len(self._tag_stack):            if self._tag_stack[-1] == 'ol':                self._append_to_markdown('1.    ' + self._handled_tag_body_data + os.linesep)            elif self._tag_stack[-1] == 'ul':                self._append_to_markdown('*    ' + self._handled_tag_body_data + os.linesep)    # <p />    def handle_start_p(self, attrs):        if len(self._markdown) > 1:            if self._markdown[-2:] == '%s%s' % (os.linesep, os.linesep):                pass            elif self._markdown[-1:] == os.linesep:                self._markdown += os.linesep            else:                self._markdown += os.linesep + os.linesep    def handle_end_p(self):        self._markdown += '%s%s' % (os.linesep, os.linesep)    # <pre />    def handle_end_pre(self):        code_lines = self._handled_tag_body_data.split('\n')        for code_line in code_lines:            self._append_to_markdown('    ' + code_line + os.linesep)    # <strong />    def handle_end_strong(self):        self._handled_tag_body_data = self._handled_tag_body_data.replace(os.linesep, ' ')        self._append_to_markdown('**' + self._handled_tag_body_data + '**')    ## ###    def handle_starttag(self, tag, attrs):        self._tag_stack.append(tag)        try:            eval('self.handle_start_' + tag + '(attrs)')        except AttributeError, e:            pass    def handle_endtag(self, tag):        self._tag_stack.pop()        try:            eval('self.handle_end_' + tag + '()')            # Collapse three successive CRs into two before moving on            while len(self._markdown) > 2 and\                  self._markdown[-3:] == '%s%s%s' % (os.linesep, os.linesep, os.linesep):                self._markdown = self._markdown[:-3] + '%s%s' % (os.linesep, os.linesep)        except AttributeError, e:            pass        self._tag_attr_data = {}        self._handled_tag_body_data = ''    def handle_data(self, data):        data = os.linesep.join(data.strip().split(os.linesep))        if len(self._tag_stack) and self._tag_stack[-1] not in ['p']:            self._handled_tag_body_data += data        else:            self._append_to_markdown(data)    def get_markdown(self):        return self._markdown.rstrip() + '\n'def main():    p = Html2MarkdownParser()    buf = ''' here place your html code'''    p.feed(buf)    p.close()    print p.get_markdown()if __name__ == "__main__":    sys.exit(main())

注意这个只可以做有限的转换，table，div等标签是转换不了的。