diff --git a/ebook_converter/ebooks/textile/__init__.py b/ebook_converter/ebooks/textile/__init__.py new file mode 100644 index 0000000..5a0a767 --- /dev/null +++ b/ebook_converter/ebooks/textile/__init__.py @@ -0,0 +1,7 @@ +from __future__ import absolute_import, division, print_function, unicode_literals +from .functions import textile, textile_restricted, Textile + +if False: + textile, textile_restricted, Textile + +__all__ = ['textile', 'textile_restricted'] diff --git a/ebook_converter/ebooks/textile/functions.py b/ebook_converter/ebooks/textile/functions.py new file mode 100644 index 0000000..5029a2a --- /dev/null +++ b/ebook_converter/ebooks/textile/functions.py @@ -0,0 +1,1091 @@ +#!/usr/bin/env python2 +# -*- coding: utf-8 -*- +from __future__ import absolute_import, division, print_function, unicode_literals + +""" +PyTextile + +A Humane Web Text Generator +""" + +# Last upstream version basis +# __version__ = '2.1.4' +# __date__ = '2009/12/04' + +__copyright__ = """ +Copyright (c) 2011, Leigh Parry +Copyright (c) 2011, John Schember +Copyright (c) 2009, Jason Samsa, http://jsamsa.com/ +Copyright (c) 2004, Roberto A. F. De Almeida, http://dealmeida.net/ +Copyright (c) 2003, Mark Pilgrim, http://diveintomark.org/ + +Original PHP Version: +Copyright (c) 2003-2004, Dean Allen +All rights reserved. + +Thanks to Carlo Zottmann for refactoring +Textile's procedural code into a class framework + +Additions and fixes Copyright (c) 2006 Alex Shiels http://thresholdstate.com/ + +""" + +__license__ = """ +L I C E N S E +============= +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, + this list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name Textile nor the names of its contributors may be used to + endorse or promote products derived from this software without specific + prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE +ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE +LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR +CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF +SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS +INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN +CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) +ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE +POSSIBILITY OF SUCH DAMAGE. + +""" + +import re +import uuid + +from calibre.utils.smartypants import smartyPants +from polyglot.builtins import unicode_type +from polyglot.urllib import urlopen, urlparse + + +def _normalize_newlines(string): + out = re.sub(r'\r\n', '\n', string) + out = re.sub(r'\n{3,}', '\n\n', out) + out = re.sub(r'\n\s*\n', '\n\n', out) + out = re.sub(r'"$', '" ', out) + return out + + +def getimagesize(url): + """ + Attempts to determine an image's width and height, and returns a string + suitable for use in an tag, or None in case of failure. + Requires that PIL is installed. + + >>> getimagesize("http://www.google.com/intl/en_ALL/images/logo.gif") + ... #doctest: +ELLIPSIS, +SKIP + 'width="..." height="..."' + + """ + + from PIL import ImageFile + + try: + p = ImageFile.Parser() + f = urlopen(url) + while True: + s = f.read(1024) + if not s: + break + p.feed(s) + if p.image: + return 'width="%i" height="%i"' % p.image.size + except (IOError, ValueError): + return None + + +class Textile(object): + hlgn = r'(?:\<(?!>)|(?|\<\>|\=|[()]+(?! ))' + vlgn = r'[\-^~]' + clas = r'(?:\([^)]+\))' + lnge = r'(?:\[[^\]]+\])' + styl = r'(?:\{[^}]+\})' + cspn = r'(?:\\\d+)' + rspn = r'(?:\/\d+)' + a = r'(?:%s|%s)*' % (hlgn, vlgn) + s = r'(?:%s|%s)*' % (cspn, rspn) + c = r'(?:%s)*' % '|'.join([clas, styl, lnge, hlgn]) + + pnct = r'[-!"#$%&()*+,/:;<=>?@\'\[\\\]\.^_`{|}~]' + # urlch = r'[\w"$\-_.+!*\'(),";/?:@=&%#{}|\\^~\[\]`]' + urlch = r'[\w"$\-_.+*\'(),";\/?:@=&%#{}|\\^~\[\]`]' + + url_schemes = ('http', 'https', 'ftp', 'mailto') + + btag = ('bq', 'bc', 'notextile', 'pre', 'h[1-6]', r'fn\d+', 'p') + btag_lite = ('bq', 'bc', 'p') + + macro_defaults = [ + (re.compile(r'{(c\||\|c)}'), r'¢'), # cent + (re.compile(r'{(L-|-L)}'), r'£'), # pound + (re.compile(r'{(Y=|=Y)}'), r'¥'), # yen + (re.compile(r'{\(c\)}'), r'©'), # copyright + (re.compile(r'{\(r\)}'), r'®'), # registered + (re.compile(r'{(\+_|_\+)}'), r'±'), # plus-minus + (re.compile(r'{1/4}'), r'¼'), # quarter + (re.compile(r'{1/2}'), r'½'), # half + (re.compile(r'{3/4}'), r'¾'), # three-quarter + (re.compile(r'{(A`|`A)}'), r'À'), # A-acute + (re.compile(r'{(A\'|\'A)}'), r'Á'), # A-grave + (re.compile(r'{(A\^|\^A)}'), r'Â'), # A-circumflex + (re.compile(r'{(A~|~A)}'), r'Ã'), # A-tilde + (re.compile(r'{(A\"|\"A)}'), r'Ä'), # A-diaeresis + (re.compile(r'{(Ao|oA)}'), r'Å'), # A-ring + (re.compile(r'{(AE)}'), r'Æ'), # AE + (re.compile(r'{(C,|,C)}'), r'Ç'), # C-cedilla + (re.compile(r'{(E`|`E)}'), r'È'), # E-acute + (re.compile(r'{(E\'|\'E)}'), r'É'), # E-grave + (re.compile(r'{(E\^|\^E)}'), r'Ê'), # E-circumflex + (re.compile(r'{(E\"|\"E)}'), r'Ë'), # E-diaeresis + (re.compile(r'{(I`|`I)}'), r'Ì'), # I-acute + (re.compile(r'{(I\'|\'I)}'), r'Í'), # I-grave + (re.compile(r'{(I\^|\^I)}'), r'Î'), # I-circumflex + (re.compile(r'{(I\"|\"I)}'), r'Ï'), # I-diaeresis + (re.compile(r'{(D-|-D)}'), r'Ð'), # ETH + (re.compile(r'{(N~|~N)}'), r'Ñ'), # N-tilde + (re.compile(r'{(O`|`O)}'), r'Ò'), # O-acute + (re.compile(r'{(O\'|\'O)}'), r'Ó'), # O-grave + (re.compile(r'{(O\^|\^O)}'), r'Ô'), # O-circumflex + (re.compile(r'{(O~|~O)}'), r'Õ'), # O-tilde + (re.compile(r'{(O\"|\"O)}'), r'Ö'), # O-diaeresis + (re.compile(r'{x}'), r'×'), # dimension + (re.compile(r'{(O\/|\/O)}'), r'Ø'), # O-slash + (re.compile(r'{(U`|`U)}'), r'Ù'), # U-acute + (re.compile(r'{(U\'|\'U)}'), r'Ú'), # U-grave + (re.compile(r'{(U\^|\^U)}'), r'Û'), # U-circumflex + (re.compile(r'{(U\"|\"U)}'), r'Ü'), # U-diaeresis + (re.compile(r'{(Y\'|\'Y)}'), r'Ý'), # Y-grave + (re.compile(r'{sz}'), r'ß'), # sharp-s + (re.compile(r'{(a`|`a)}'), r'à'), # a-grave + (re.compile(r'{(a\'|\'a)}'), r'á'), # a-acute + (re.compile(r'{(a\^|\^a)}'), r'â'), # a-circumflex + (re.compile(r'{(a~|~a)}'), r'ã'), # a-tilde + (re.compile(r'{(a\"|\"a)}'), r'ä'), # a-diaeresis + (re.compile(r'{(ao|oa)}'), r'å'), # a-ring + (re.compile(r'{ae}'), r'æ'), # ae + (re.compile(r'{(c,|,c)}'), r'ç'), # c-cedilla + (re.compile(r'{(e`|`e)}'), r'è'), # e-grave + (re.compile(r'{(e\'|\'e)}'), r'é'), # e-acute + (re.compile(r'{(e\^|\^e)}'), r'ê'), # e-circumflex + (re.compile(r'{(e\"|\"e)}'), r'ë'), # e-diaeresis + (re.compile(r'{(i`|`i)}'), r'ì'), # i-grave + (re.compile(r'{(i\'|\'i)}'), r'í'), # i-acute + (re.compile(r'{(i\^|\^i)}'), r'î'), # i-circumflex + (re.compile(r'{(i\"|\"i)}'), r'ï'), # i-diaeresis + (re.compile(r'{(d-|-d)}'), r'ð'), # eth + (re.compile(r'{(n~|~n)}'), r'ñ'), # n-tilde + (re.compile(r'{(o`|`o)}'), r'ò'), # o-grave + (re.compile(r'{(o\'|\'o)}'), r'ó'), # o-acute + (re.compile(r'{(o\^|\^o)}'), r'ô'), # o-circumflex + (re.compile(r'{(o~|~o)}'), r'õ'), # o-tilde + (re.compile(r'{(o\"|\"o)}'), r'ö'), # o-diaeresis + (re.compile(r'{(o\/|\/o)}'), r'ø'), # o-stroke + (re.compile(r'{(u`|`u)}'), r'ù'), # u-grave + (re.compile(r'{(u\'|\'u)}'), r'ú'), # u-acute + (re.compile(r'{(u\^|\^u)}'), r'û'), # u-circumflex + (re.compile(r'{(u\"|\"u)}'), r'ü'), # u-diaeresis + (re.compile(r'{(y\'|\'y)}'), r'ý'), # y-acute + (re.compile(r'{(y\"|\"y)}'), r'ÿ'), # y-diaeresis + + (re.compile(r'{(C\ˇ|\ˇC)}'), r'Č'), # C-caron + (re.compile(r'{(c\ˇ|\ˇc)}'), r'č'), # c-caron + (re.compile(r'{(D\ˇ|\ˇD)}'), r'Ď'), # D-caron + (re.compile(r'{(d\ˇ|\ˇd)}'), r'ď'), # d-caron + (re.compile(r'{(E\ˇ|\ˇE)}'), r'Ě'), # E-caron + (re.compile(r'{(e\ˇ|\ˇe)}'), r'ě'), # e-caron + (re.compile(r'{(L\'|\'L)}'), r'Ĺ'), # L-acute + (re.compile(r'{(l\'|\'l)}'), r'ĺ'), # l-acute + (re.compile(r'{(L\ˇ|\ˇL)}'), r'Ľ'), # L-caron + (re.compile(r'{(l\ˇ|\ˇl)}'), r'ľ'), # l-caron + (re.compile(r'{(N\ˇ|\ˇN)}'), r'Ň'), # N-caron + (re.compile(r'{(n\ˇ|\ˇn)}'), r'ň'), # n-caron + + (re.compile(r'{OE}'), r'Œ'), # OE + (re.compile(r'{oe}'), r'œ'), # oe + + (re.compile(r'{(R\'|\'R)}'), r'Ŕ'), # R-acute + (re.compile(r'{(r\'|\'r)}'), r'ŕ'), # r-acute + (re.compile(r'{(R\ˇ|\ˇR)}'), r'Ř'), # R-caron + (re.compile(r'{(r\ˇ|\ˇr)}'), r'ř'), # r-caron + + (re.compile(r'{(S\^|\^S)}'), r'Ŝ'), # S-circumflex + (re.compile(r'{(s\^|\^s)}'), r'ŝ'), # s-circumflex + + (re.compile(r'{(S\ˇ|\ˇS)}'), r'Š'), # S-caron + (re.compile(r'{(s\ˇ|\ˇs)}'), r'š'), # s-caron + (re.compile(r'{(T\ˇ|\ˇT)}'), r'Ť'), # T-caron + (re.compile(r'{(t\ˇ|\ˇt)}'), r'ť'), # t-caron + (re.compile(r'{(U\°|\°U)}'), r'Ů'), # U-ring + (re.compile(r'{(u\°|\°u)}'), r'ů'), # u-ring + (re.compile(r'{(Z\ˇ|\ˇZ)}'), r'Ž'), # Z-caron + (re.compile(r'{(z\ˇ|\ˇz)}'), r'ž'), # z-caron + + (re.compile(r'{\*}'), r'•'), # bullet + (re.compile(r'{Fr}'), r'₣'), # Franc + (re.compile(r'{(L=|=L)}'), r'₤'), # Lira + (re.compile(r'{Rs}'), r'₨'), # Rupee + (re.compile(r'{(C=|=C)}'), r'€'), # euro + (re.compile(r'{tm}'), r'™'), # trademark + (re.compile(r'{spades?}'), r'♠'), # spade + (re.compile(r'{clubs?}'), r'♣'), # club + (re.compile(r'{hearts?}'), r'♥'), # heart + (re.compile(r'{diam(onds?|s)}'), r'♦'), # diamond + (re.compile(r'{"}'), r'"'), # double-quote + (re.compile(r"{'}"), r'''), # single-quote + (re.compile(r"{(’|'/|/')}"), r'’'), # closing-single-quote - apostrophe + (re.compile(r"{(‘|\\'|'\\)}"), r'‘'), # opening-single-quote + (re.compile(r'{(”|"/|/")}'), r'”'), # closing-double-quote + (re.compile(r'{(“|\\"|"\\)}'), r'“'), # opening-double-quote + ] + glyph_defaults = [ + (re.compile(r'(\d+\'?\"?)( ?)x( ?)(?=\d+)'), r'\1\2×\3'), # dimension sign + (re.compile(r'(\d+)\'(\s)', re.I), r'\1′\2'), # prime + (re.compile(r'(\d+)\"(\s)', re.I), r'\1″\2'), # prime-double + (re.compile(r'\b([A-Z][A-Z0-9]{2,})\b(?:[(]([^)]*)[)])'), r'\1'), # 3+ uppercase acronym + (re.compile(r'\b([A-Z][A-Z\'\-]+[A-Z])(?=[\s.,\)>])'), r'\1'), # 3+ uppercase + (re.compile(r'\b(\s{0,1})?\.{3}'), r'\1…'), # ellipsis + (re.compile(r'^[\*_-]{3,}$', re.M), r'
'), #
scene-break + (re.compile(r'(^|[^-])--([^-]|$)'), r'\1—\2'), # em dash + (re.compile(r'\s-(?:\s|$)'), r' – '), # en dash + (re.compile(r'\b( ?)[([]TM[])]', re.I), r'\1™'), # trademark + (re.compile(r'\b( ?)[([]R[])]', re.I), r'\1®'), # registered + (re.compile(r'\b( ?)[([]C[])]', re.I), r'\1©'), # copyright + ] + + def __init__(self, restricted=False, lite=False, noimage=False): + """docstring for __init__""" + self.restricted = restricted + self.lite = lite + self.noimage = noimage + self.get_sizes = False + self.fn = {} + self.urlrefs = {} + self.shelf = {} + self.rel = '' + self.html_type = 'xhtml' + + def textile(self, text, rel=None, head_offset=0, html_type='xhtml'): + """ + >>> import textile + >>> textile.textile('some textile') + u'\\t

some textile

' + """ + self.html_type = html_type + + # text = type(u'')(text) + text = _normalize_newlines(text) + + if self.restricted: + text = self.encode_html(text, quotes=False) + + if rel: + self.rel = ' rel="%s"' % rel + + text = self.getRefs(text) + text = self.block(text, int(head_offset)) + text = self.retrieve(text) + text = smartyPants(text, 'q') + + return text + + def pba(self, input, element=None): + """ + Parse block attributes. + + >>> t = Textile() + >>> t.pba(r'\3') + '' + >>> t.pba(r'\\3', element='td') + ' colspan="3"' + >>> t.pba(r'/4', element='td') + ' rowspan="4"' + >>> t.pba(r'\\3/4', element='td') + ' colspan="3" rowspan="4"' + + >>> t.vAlign('^') + 'top' + + >>> t.pba('^', element='td') + ' style="vertical-align:top;"' + + >>> t.pba('{line-height:18px}') + ' style="line-height:18px;"' + + >>> t.pba('(foo-bar)') + ' class="foo-bar"' + + >>> t.pba('(#myid)') + ' id="myid"' + + >>> t.pba('(foo-bar#myid)') + ' class="foo-bar" id="myid"' + + >>> t.pba('((((') + ' style="padding-left:4em;"' + + >>> t.pba(')))') + ' style="padding-right:3em;"' + + >>> t.pba('[fr]') + ' lang="fr"' + + """ + style = [] + aclass = '' + lang = '' + colspan = '' + rowspan = '' + id = '' + + if not input: + return '' + + matched = input + if element == 'td': + m = re.search(r'\\(\d+)', matched) + if m: + colspan = m.group(1) + + m = re.search(r'/(\d+)', matched) + if m: + rowspan = m.group(1) + + if element == 'td' or element == 'tr': + m = re.search(r'(%s)' % self.vlgn, matched) + if m: + style.append("vertical-align:%s;" % self.vAlign(m.group(1))) + + m = re.search(r'\{([^}]*)\}', matched) + if m: + style.append(m.group(1).rstrip(';') + ';') + matched = matched.replace(m.group(0), '') + + m = re.search(r'\[([^\]]+)\]', matched, re.U) + if m: + lang = m.group(1) + matched = matched.replace(m.group(0), '') + + m = re.search(r'\(([^()]+)\)', matched, re.U) + if m: + aclass = m.group(1) + matched = matched.replace(m.group(0), '') + + m = re.search(r'([(]+)', matched) + if m: + style.append("padding-left:%sem;" % len(m.group(1))) + matched = matched.replace(m.group(0), '') + + m = re.search(r'([)]+)', matched) + if m: + style.append("padding-right:%sem;" % len(m.group(1))) + matched = matched.replace(m.group(0), '') + + m = re.search(r'(%s)' % self.hlgn, matched) + if m: + style.append("text-align:%s;" % self.hAlign(m.group(1))) + + m = re.search(r'^(.*)#(.*)$', aclass) + if m: + id = m.group(2) + aclass = m.group(1) + + if self.restricted: + if lang: + return ' lang="%s"' + else: + return '' + + result = [] + if style: + result.append(' style="%s"' % "".join(style)) + if aclass: + result.append(' class="%s"' % aclass) + if lang: + result.append(' lang="%s"' % lang) + if id: + result.append(' id="%s"' % id) + if colspan: + result.append(' colspan="%s"' % colspan) + if rowspan: + result.append(' rowspan="%s"' % rowspan) + return ''.join(result) + + def hasRawText(self, text): + """ + checks whether the text has text not already enclosed by a block tag + + >>> t = Textile() + >>> t.hasRawText('

foo bar biz baz

') + False + + >>> t.hasRawText(' why yes, yes it does') + True + + """ + r = re.compile(r'<(p|blockquote|div|form|table|ul|ol|pre|h\d)[^>]*?>.*', re.S).sub('', text.strip()).strip() + r = re.compile(r'<(hr|br)[^>]*?/>').sub('', r) + return '' != r + + def table(self, text): + r""" + >>> t = Textile() + >>> t.table('|one|two|three|\n|a|b|c|') + '\t\n\t\t\n\t\t\t\n\t\t\t\n\t\t\t\n\t\t\n\t\t\n\t\t\t\n\t\t\t\n\t\t\t\n\t\t\n\t
onetwothree
abc
\n\n' + """ + text = text + "\n\n" + pattern = re.compile(r'^(?:table(_?%(s)s%(a)s%(c)s)\. ?\n)?^(%(a)s%(c)s\.? ?\|.*\|)\n\n' % {'s':self.s, 'a':self.a, 'c':self.c}, re.S|re.M|re.U) + return pattern.sub(self.fTable, text) + + def fTable(self, match): + tatts = self.pba(match.group(1), 'table') + rows = [] + for row in [x for x in match.group(2).split('\n') if x]: + rmtch = re.search(r'^(%s%s\. )(.*)' % (self.a, self.c), row.lstrip()) + if rmtch: + ratts = self.pba(rmtch.group(1), 'tr') + row = rmtch.group(2) + else: + ratts = '' + + cells = [] + for cell in row.split('|')[1:-1]: + ctyp = 'd' + if re.search(r'^_', cell): + ctyp = "h" + cmtch = re.search(r'^(_?%s%s%s\. )(.*)' % (self.s, self.a, self.c), cell) + if cmtch: + catts = self.pba(cmtch.group(1), 'td') + cell = cmtch.group(2) + else: + catts = '' + + cell = self.graf(self.span(cell)) + cells.append('\t\t\t%s' % (ctyp, catts, cell, ctyp)) + rows.append("\t\t\n%s\n\t\t" % (ratts, '\n'.join(cells))) + cells = [] + catts = None + return "\t\n%s\n\t\n\n" % (tatts, '\n'.join(rows)) + + def lists(self, text): + """ + >>> t = Textile() + >>> t.lists("* one\\n* two\\n* three") + '\\t
    \\n\\t\\t
  • one
    • \\n\\t\\t
  • two
    • \\n\\t\\t
  • three
    • \\n\\t
' + """ + pattern = re.compile(r'^([#*]+%s .*)$(?![^#*])' % self.c, re.U|re.M|re.S) + return pattern.sub(self.fList, text) + + def fList(self, match): + text = match.group(0).split("\n") + result = [] + lists = [] + for i, line in enumerate(text): + try: + nextline = text[i+1] + except IndexError: + nextline = '' + + m = re.search(r"^([#*]+)(%s%s) (.*)$" % (self.a, self.c), line, re.S) + if m: + tl, atts, content = m.groups() + nl = '' + nm = re.search(r'^([#*]+)\s.*', nextline) + if nm: + nl = nm.group(1) + if tl not in lists: + lists.append(tl) + atts = self.pba(atts) + line = "\t<%sl%s>\n\t\t
  • %s" % (self.lT(tl), atts, self.graf(content)) + else: + line = "\t\t
  • " + self.graf(content) + + if len(nl) <= len(tl): + line = line + "
    • " + for k in reversed(lists): + if len(k) > len(nl): + line = line + "\n\t" % self.lT(k) + if len(k) > 1: + line = line + "" + lists.remove(k) + + result.append(line) + return "\n".join(result) + + def lT(self, input): + if re.search(r'^#+', input): + return 'o' + else: + return 'u' + + def doPBr(self, in_): + return re.compile(r'<(p)([^>]*?)>(.*)()', re.S).sub(self.doBr, in_) + + def doBr(self, match): + if self.html_type == 'html': + content = re.sub(r'(.+)(?:(?)|(?))\n(?![#*\s|])', '\\1
    ', match.group(3)) + else: + content = re.sub(r'(.+)(?:(?)|(?))\n(?![#*\s|])', '\\1
    ', match.group(3)) + return '<%s%s>%s%s' % (match.group(1), match.group(2), content, match.group(4)) + + def block(self, text, head_offset=0): + """ + >>> t = Textile() + >>> t.block('h1. foobar baby') + '\\t

    foobar baby

    ' + """ + if not self.lite: + tre = '|'.join(self.btag) + else: + tre = '|'.join(self.btag_lite) + text = text.split('\n\n') + + tag = 'p' + atts = cite = graf = ext = c1 = '' + + out = [] + + anon = False + for line in text: + pattern = r'^(%s)(%s%s)\.(\.?)(?::(\S+))? (.*)$' % (tre, self.a, self.c) + match = re.search(pattern, line, re.S) + if match: + if ext: + out.append(out.pop() + c1) + + tag, atts, ext, cite, graf = match.groups() + h_match = re.search(r'h([1-6])', tag) + if h_match: + head_level, = h_match.groups() + tag = 'h%i' % max(1, + min(int(head_level) + head_offset, + 6)) + o1, o2, content, c2, c1 = self.fBlock(tag, atts, ext, + cite, graf) + # leave off c1 if this block is extended, + # we'll close it at the start of the next block + + if ext: + line = "%s%s%s%s" % (o1, o2, content, c2) + else: + line = "%s%s%s%s%s" % (o1, o2, content, c2, c1) + + else: + anon = True + if ext or not re.search(r'^\s', line): + o1, o2, content, c2, c1 = self.fBlock(tag, atts, ext, + cite, line) + # skip $o1/$c1 because this is part of a continuing + # extended block + if tag == 'p' and not self.hasRawText(content): + line = content + else: + line = "%s%s%s" % (o2, content, c2) + else: + line = self.graf(line) + + line = self.doPBr(line) + if self.html_type == 'xhtml': + line = re.sub(r'
    ', '
    ', line) + + if ext and anon: + out.append(out.pop() + "\n" + line) + else: + out.append(line) + + if not ext: + tag = 'p' + atts = '' + cite = '' + graf = '' + + if ext: + out.append(out.pop() + c1) + return '\n\n'.join(out) + + def fBlock(self, tag, atts, ext, cite, content): + """ + >>> t = Textile() + >>> t.fBlock("bq", "", None, "", "Hello BlockQuote") + ('\\t
    \\n', '\\t\\t

    ', 'Hello BlockQuote', '

    ', '\\n\\t
    ') + + >>> t.fBlock("bq", "", None, "http://google.com", "Hello BlockQuote") + ('\\t
    \\n', '\\t\\t

    ', 'Hello BlockQuote', '

    ', '\\n\\t
    ') + + >>> t.fBlock("bc", "", None, "", 'printf "Hello, World";') # doctest: +ELLIPSIS + ('
    ', '', ..., '', '
    ') + + >>> t.fBlock("h1", "", None, "", "foobar") + ('', '\\t

    ', 'foobar', '

    ', '') + """ + atts = self.pba(atts) + o1 = o2 = c2 = c1 = '' + + m = re.search(r'fn(\d+)', tag) + if m: + tag = 'p' + if m.group(1) in self.fn: + fnid = self.fn[m.group(1)] + else: + fnid = m.group(1) + atts = atts + ' id="fn%s"' % fnid + if atts.find('class=') < 0: + atts = atts + ' class="footnote"' + content = ('%s' % m.group(1)) + content + + if tag == 'bq': + cite = self.checkRefs(cite) + if cite: + cite = ' cite="%s"' % cite + else: + cite = '' + o1 = "\t\n" % (cite, atts) + o2 = "\t\t" % atts + c2 = "

    " + c1 = "\n\t" + + elif tag == 'bc': + o1 = "" % atts + o2 = "" % atts + c2 = "" + c1 = "" + content = self.shelve(self.encode_html(content.rstrip("\n") + "\n")) + + elif tag == 'notextile': + content = self.shelve(content) + o1 = o2 = '' + c1 = c2 = '' + + elif tag == 'pre': + content = self.shelve(self.encode_html(content.rstrip("\n") + "\n")) + o1 = "" % atts + o2 = c2 = '' + c1 = '' + + else: + o2 = "\t<%s%s>" % (tag, atts) + c2 = "" % tag + + content = self.graf(content) + return o1, o2, content, c2, c1 + + def footnoteRef(self, text): + """ + >>> t = Textile() + >>> t.footnoteRef('foo[1] ') # doctest: +ELLIPSIS + 'foo1 ' + """ + return re.sub(r'\b\[([0-9]+)\](\s)?', self.footnoteID, text) + + def footnoteID(self, match): + id, t = match.groups() + if id not in self.fn: + self.fn[id] = unicode_type(uuid.uuid4()) + fnid = self.fn[id] + if not t: + t = '' + return '%s%s' % (fnid, id, t) + + def glyphs(self, text): + """ + >>> t = Textile() + + >>> t.glyphs("apostrophe's") + 'apostrophe’s' + + >>> t.glyphs("back in '88") + 'back in ’88' + + >>> t.glyphs('foo ...') + 'foo …' + + >>> t.glyphs('--') + '—' + + >>> t.glyphs('FooBar[tm]') + 'FooBar™' + + >>> t.glyphs("

    Cat's Cradle by Vonnegut

    ") + '

    Cat’s Cradle by Vonnegut

    ' + + """ + # fix: hackish + text = re.sub(r'"\Z', '\" ', text) + + result = [] + for line in re.compile(r'(<.*?>)', re.U).split(text): + if not re.search(r'<.*>', line): + rules = [] + if re.search(r'{.+?}', line): + rules = self.macro_defaults + self.glyph_defaults + else: + rules = self.glyph_defaults + for s, r in rules: + line = s.sub(r, line) + result.append(line) + return ''.join(result) + + def macros_only(self, text): + # fix: hackish + text = re.sub(r'"\Z', '\" ', text) + + result = [] + for line in re.compile(r'(<.*?>)', re.U).split(text): + if not re.search(r'<.*>', line): + rules = [] + if re.search(r'{.+?}', line): + rules = self.macro_defaults + for s, r in rules: + line = s.sub(r, line) + result.append(line) + return ''.join(result) + + def vAlign(self, input): + d = {'^':'top', '-':'middle', '~':'bottom'} + return d.get(input, '') + + def hAlign(self, input): + d = {'<':'left', '=':'center', '>':'right', '<>': 'justify'} + return d.get(input, '') + + def getRefs(self, text): + """ + what is this for? + """ + pattern = re.compile(r'(?:(?<=^)|(?<=\s))\[(.+)\]((?:http(?:s?):\/\/|\/)\S+)(?=\s|$)', re.U) + text = pattern.sub(self.refs, text) + return text + + def refs(self, match): + flag, url = match.groups() + self.urlrefs[flag] = url + return '' + + def checkRefs(self, url): + return self.urlrefs.get(url, url) + + def isRelURL(self, url): + """ + Identify relative urls. + + >>> t = Textile() + >>> t.isRelURL("http://www.google.com/") + False + >>> t.isRelURL("/foo") + True + + """ + (scheme, netloc) = urlparse(url)[0:2] + return not scheme and not netloc + + def relURL(self, url): + scheme = urlparse(url)[0] + if self.restricted and scheme and scheme not in self.url_schemes: + return '#' + return url + + def shelve(self, text): + id = unicode_type(uuid.uuid4()) + 'c' + self.shelf[id] = text + return id + + def retrieve(self, text): + """ + >>> t = Textile() + >>> id = t.shelve("foobar") + >>> t.retrieve(id) + 'foobar' + """ + while True: + old = text + for k, v in self.shelf.items(): + text = text.replace(k, v) + if text == old: + break + return text + + def encode_html(self, text, quotes=True): + a = ( + ('&', '&'), + ('<', '<'), + ('>', '>') + ) + + if quotes: + a = a + ( + ("'", '''), + ('"', '"') + ) + + for k, v in a: + text = text.replace(k, v) + return text + + def graf(self, text): + if not self.lite: + text = self.noTextile(text) + text = self.code(text) + + text = self.links(text) + + if not self.noimage: + text = self.image(text) + + if not self.lite: + text = self.lists(text) + text = self.table(text) + + text = self.span(text) + text = self.footnoteRef(text) + text = self.glyphs(text) + + return text.rstrip('\n') + + def links(self, text): + """ + >>> t = Textile() + >>> t.links('fooobar "Google":http://google.com/foobar/ and hello world "flickr":http://flickr.com/photos/jsamsa/ ') # doctest: +ELLIPSIS + 'fooobar ... and hello world ...' + """ + + text = self.macros_only(text) + punct = '!"#$%&\'*+,-./:;=?@\\^_`|~' + + pattern = r''' + (?P
        [\s\[{(]|[%s]   )?
+            "                          # start
+            (?P   %s       )
+            (?P   [^"]+?   )
+            \s?
+            (?:   \(([^)]+?)\)(?=")   )?     # $title
+            ":
+            (?P    (?:ftp|https?)? (?: :// )? [-A-Za-z0-9+&@#/?=~_()|!:,.;]*[-A-Za-z0-9+&@#/=~_()|]   )
+            (?P   [^\w\/;]*?   )
+            (?=<|\s|$)
+        ''' % (re.escape(punct), self.c)
+
+        text = re.compile(pattern, re.X).sub(self.fLink, text)
+
+        return text
+
+    def fLink(self, match):
+        pre, atts, text, title, url, post = match.groups()
+
+        if pre is None:
+            pre = ''
+
+        # assume ) at the end of the url is not actually part of the url
+        # unless the url also contains a (
+        if url.endswith(')') and not url.find('(') > -1:
+            post = url[-1] + post
+            url = url[:-1]
+
+        url = self.checkRefs(url)
+
+        atts = self.pba(atts)
+        if title:
+            atts = atts +  ' title="%s"' % self.encode_html(title)
+
+        if not self.noimage:
+            text = self.image(text)
+
+        text = self.span(text)
+        text = self.glyphs(text)
+
+        url = self.relURL(url)
+        out = '%s' % (self.encode_html(url), atts, self.rel, text)
+        out = self.shelve(out)
+        return ''.join([pre, out, post])
+
+    def span(self, text):
+        """
+        >>> t = Textile()
+        >>> t.span(r"hello %(bob)span *strong* and **bold**% goodbye")
+        'hello span strong and bold goodbye'
+        """
+        qtags = (r'\*\*', r'\*', r'\?\?', r'\-', r'__', r'_', r'%', r'\+', r'~', r'\^')
+        pnct = ".,\"'?!;:"
+
+        for qtag in qtags:
+            pattern = re.compile(r"""
+                (?:^|(?<=[\s>%(pnct)s\(])|\[|([\]}]))
+                (%(qtag)s)(?!%(qtag)s)
+                (%(c)s)
+                (?::(\S+))?
+                ([^\s%(qtag)s]+|\S[^%(qtag)s\n]*[^\s%(qtag)s\n])
+                ([%(pnct)s]*)
+                %(qtag)s
+                (?:$|([\]}])|(?=%(selfpnct)s{1,2}|\s))
+            """ % {'qtag':qtag, 'c':self.c, 'pnct':pnct,
+                   'selfpnct':self.pnct}, re.X)
+            text = pattern.sub(self.fSpan, text)
+        return text
+
+    def fSpan(self, match):
+        _, tag, atts, cite, content, end, _ = match.groups()
+
+        qtags = {
+            '*': 'strong',
+            '**': 'b',
+            '??': 'cite',
+            '_' : 'em',
+            '__': 'i',
+            '-' : 'del',
+            '%' : 'span',
+            '+' : 'ins',
+            '~' : 'sub',
+            '^' : 'sup'
+        }
+        tag = qtags[tag]
+        atts = self.pba(atts)
+        if cite:
+            atts = atts + 'cite="%s"' % cite
+
+        content = self.span(content)
+
+        out = "<%s%s>%s%s" % (tag, atts, content, end, tag)
+        return out
+
+    def image(self, text):
+        """
+        >>> t = Textile()
+        >>> t.image('!/imgs/myphoto.jpg!:http://jsamsa.com')
+        ''
+        """
+        pattern = re.compile(r"""
+            (?:[\[{])?          # pre
+            \!                 # opening !
+            (%s)               # optional style,class atts
+            (?:\. )?           # optional dot-space
+            ([^\s(!]+)         # presume this is the src
+            \s?                # optional space
+            (?:\(([^\)]+)\))?  # optional title
+            \!                 # closing
+            (?::(\S+))?        # optional href
+            (?:[\]}]|(?=\s|$)) # lookahead: space or end of string
+        """ % self.c, re.U|re.X)
+        return pattern.sub(self.fImage, text)
+
+    def fImage(self, match):
+        # (None, '', '/imgs/myphoto.jpg', None, None)
+        atts, url, title, href = match.groups()
+        atts  = self.pba(atts)
+
+        if title:
+            atts = atts + ' title="%s" alt="%s"' % (title, title)
+        else:
+            atts = atts + ' alt=""'
+
+        if not self.isRelURL(url) and self.get_sizes:
+            size = getimagesize(url)
+            if (size):
+                atts += " %s" % size
+
+        if href:
+            href = self.checkRefs(href)
+
+        url = self.checkRefs(url)
+        url = self.relURL(url)
+
+        out = []
+        if href:
+            out.append('' % href)
+        if self.html_type == 'html':
+            out.append('' % (url, atts))
+        else:
+            out.append('' % (url, atts))
+        if href:
+            out.append('')
+
+        return ''.join(out)
+
+    def code(self, text):
+        text = self.doSpecial(text, '', '', self.fCode)
+        text = self.doSpecial(text, '@', '@', self.fCode)
+        text = self.doSpecial(text, '
    ', '
    ', self.fPre)
+        return text
+
+    def fCode(self, match):
+        before, text, after = match.groups()
+        if after is None:
+            after = ''
+        # text needs to be escaped
+        if not self.restricted:
+            text = self.encode_html(text)
+        return ''.join([before, self.shelve('%s' % text), after])
+
+    def fPre(self, match):
+        before, text, after = match.groups()
+        if after is None:
+            after = ''
+        # text needs to be escapedd
+        if not self.restricted:
+            text = self.encode_html(text)
+        return ''.join([before, '
    ', self.shelve(text), '
    ', after])
+
+    def doSpecial(self, text, start, end, method=None):
+        if method is None:
+            method = self.fSpecial
+        pattern = re.compile(r'(^|\s|[\[({>])%s(.*?)%s(\s|$|[\])}])?' % (re.escape(start), re.escape(end)), re.M|re.S)
+        return pattern.sub(method, text)
+
+    def fSpecial(self, match):
+        """
+        special blocks like notextile or code
+        """
+        before, text, after = match.groups()
+        if after is None:
+            after = ''
+        return ''.join([before, self.shelve(self.encode_html(text)), after])
+
+    def noTextile(self, text):
+        text = self.doSpecial(text, '', '', self.fTextile)
+        return self.doSpecial(text, '==', '==', self.fTextile)
+
+    def fTextile(self, match):
+        before, notextile, after = match.groups()
+        if after is None:
+            after = ''
+        return ''.join([before, self.shelve(notextile), after])
+
+
+def textile(text, head_offset=0, html_type='xhtml', encoding=None, output=None):
+    """
+    this function takes additional parameters:
+    head_offset - offset to apply to heading levels (default: 0)
+    html_type - 'xhtml' or 'html' style tags (default: 'xhtml')
+    """
+    return Textile().textile(text, head_offset=head_offset,
+                             html_type=html_type)
+
+
+def textile_restricted(text, lite=True, noimage=True, html_type='xhtml'):
+    """
+    Restricted version of Textile designed for weblog comments and other
+    untrusted input.
+
+    Raw HTML is escaped.
+    Style attributes are disabled.
+    rel='nofollow' is added to external links.
+
+    When lite=True is set (the default):
+    Block tags are restricted to p, bq, and bc.
+    Lists and tables are disabled.
+
+    When noimage=True is set (the default):
+    Image tags are disabled.
+
+    """
+    return Textile(restricted=True, lite=lite,
+                   noimage=noimage).textile(text, rel='nofollow',
+                                            html_type=html_type)
diff --git a/ebook_converter/ebooks/textile/unsmarten.py b/ebook_converter/ebooks/textile/unsmarten.py
new file mode 100644
index 0000000..ff05227
--- /dev/null
+++ b/ebook_converter/ebooks/textile/unsmarten.py
@@ -0,0 +1,129 @@
+# -*- coding: utf-8 -*-
+from __future__ import absolute_import, division, print_function, unicode_literals
+
+__license__ = 'GPL 3'
+__copyright__ = '2011, Leigh Parry '
+__docformat__ = 'restructuredtext en'
+
+import re
+
+
+def unsmarten(txt):
+    txt = re.sub(u'¢|¢|¢',     r'{c\}',  txt)  # cent
+    txt = re.sub(u'£|£|£',    r'{L-}',  txt)  # pound
+    txt = re.sub(u'¥|¥|¥',      r'{Y=}',  txt)  # yen
+    txt = re.sub(u'©|©|©',     r'{(c)}', txt)  # copyright
+    txt = re.sub(u'®|®|®',      r'{(r)}', txt)  # registered
+    txt = re.sub(u'¼|¼|¼',   r'{1/4}', txt)  # quarter
+    txt = re.sub(u'½|½|½',   r'{1/2}', txt)  # half
+    txt = re.sub(u'¾|¾|¾',   r'{3/4}', txt)  # three-quarter
+    txt = re.sub(u'À|À|À',   r'{A`)}', txt)  # A-grave
+    txt = re.sub(u'Á|Á|Á',   r"{A'}",  txt)  # A-acute
+    txt = re.sub(u'Â|Â|Â',    r'{A^}',  txt)  # A-circumflex
+    txt = re.sub(u'Ã|Ã|Ã',   r'{A~}',  txt)  # A-tilde
+    txt = re.sub(u'Ä|Ä|Ä',     r'{A"}',  txt)  # A-umlaut
+    txt = re.sub(u'Å|Å|Å',    r'{Ao}',  txt)  # A-ring
+    txt = re.sub(u'Æ|Æ|Æ',    r'{AE}',  txt)  # AE
+    txt = re.sub(u'Ç|Ç|Ç',   r'{C,}',  txt)  # C-cedilla
+    txt = re.sub(u'È|È|È',   r'{E`}',  txt)  # E-grave
+    txt = re.sub(u'É|É|É',   r"{E'}",  txt)  # E-acute
+    txt = re.sub(u'Ê|Ê|Ê',    r'{E^}',  txt)  # E-circumflex
+    txt = re.sub(u'Ë|Ë|Ë',     r'{E"}',  txt)  # E-umlaut
+    txt = re.sub(u'Ì|Ì|Ì',   r'{I`}',  txt)  # I-grave
+    txt = re.sub(u'Í|Í|Í',   r"{I'}",  txt)  # I-acute
+    txt = re.sub(u'Î|Î|Î',    r'{I^}',  txt)  # I-circumflex
+    txt = re.sub(u'Ï|Ï|Ï',     r'{I"}',  txt)  # I-umlaut
+    txt = re.sub(u'Ð|Ð|Ð',      r'{D-}',  txt)  # ETH
+    txt = re.sub(u'Ñ|Ñ|Ñ',   r'{N~}',  txt)  # N-tilde
+    txt = re.sub(u'Ò|Ò|Ò',   r'{O`}',  txt)  # O-grave
+    txt = re.sub(u'Ó|Ó|Ó',   r"{O'}",  txt)  # O-acute
+    txt = re.sub(u'Ô|Ô|Ô',    r'{O^}',  txt)  # O-circumflex
+    txt = re.sub(u'Õ|Õ|Õ',   r'{O~}',  txt)  # O-tilde
+    txt = re.sub(u'Ö|Ö|Ö',     r'{O"}',  txt)  # O-umlaut
+    txt = re.sub(u'×|×|×',    r'{x}',   txt)  # dimension
+    txt = re.sub(u'Ø|Ø|Ø',   r'{O/}',  txt)  # O-slash
+    txt = re.sub(u'Ù|Ù|Ù',   r"{U`}",  txt)  # U-grave
+    txt = re.sub(u'Ú|Ú|Ú',   r"{U'}",  txt)  # U-acute
+    txt = re.sub(u'Û|Û|Û',    r'{U^}',  txt)  # U-circumflex
+    txt = re.sub(u'Ü|Ü|Ü',     r'{U"}',  txt)  # U-umlaut
+    txt = re.sub(u'Ý|Ý|Ý',   r"{Y'}",  txt)  # Y-grave
+    txt = re.sub(u'ß|ß|ß',    r'{sz}',  txt)  # sharp-s
+    txt = re.sub(u'à|à|à',   r'{a`}',  txt)  # a-grave
+    txt = re.sub(u'á|á|á',   r"{a'}",  txt)  # a-acute
+    txt = re.sub(u'â|â|â',    r'{a^}',  txt)  # a-circumflex
+    txt = re.sub(u'ã|ã|ã',   r'{a~}',  txt)  # a-tilde
+    txt = re.sub(u'ä|ä|ä',     r'{a"}',  txt)  # a-umlaut
+    txt = re.sub(u'å|å|å',    r'{ao}',  txt)  # a-ring
+    txt = re.sub(u'æ|æ|æ',    r'{ae}',  txt)  # ae
+    txt = re.sub(u'ç|ç|ç',   r'{c,}',  txt)  # c-cedilla
+    txt = re.sub(u'è|è|è',   r'{e`}',  txt)  # e-grave
+    txt = re.sub(u'é|é|é',   r"{e'}",  txt)  # e-acute
+    txt = re.sub(u'ê|ê|ê',    r'{e^}',  txt)  # e-circumflex
+    txt = re.sub(u'ë|ë|ë',     r'{e"}',  txt)  # e-umlaut
+    txt = re.sub(u'ì|ì|ì',   r'{i`}',  txt)  # i-grave
+    txt = re.sub(u'í|í|í',   r"{i'}",  txt)  # i-acute
+    txt = re.sub(u'î|î|î',    r'{i^}',  txt)  # i-circumflex
+    txt = re.sub(u'ï|ï|ï',     r'{i"}',  txt)  # i-umlaut
+    txt = re.sub(u'ð|ð|ð',      r'{d-}',  txt)  # eth
+    txt = re.sub(u'ñ|ñ|ñ',   r'{n~}',  txt)  # n-tilde
+    txt = re.sub(u'ò|ò|ò',   r'{o`}',  txt)  # o-grave
+    txt = re.sub(u'ó|ó|ó',   r"{o'}",  txt)  # o-acute
+    txt = re.sub(u'ô|ô|ô',    r'{o^}',  txt)  # o-circumflex
+    txt = re.sub(u'õ|õ|õ',   r'{o~}',  txt)  # o-tilde
+    txt = re.sub(u'ö|ö|ö',     r'{o"}',  txt)  # o-umlaut
+    txt = re.sub(u'ø|ø|ø',   r'{o/}',  txt)  # o-stroke
+    txt = re.sub(u'ù|ù|ù',   r'{u`}',  txt)  # u-grave
+    txt = re.sub(u'ú|ú|ú',   r"{u'}",  txt)  # u-acute
+    txt = re.sub(u'û|û|û',    r'{u^}',  txt)  # u-circumflex
+    txt = re.sub(u'ü|ü|ü',     r'{u"}',  txt)  # u-umlaut
+    txt = re.sub(u'ý|ý|ý',   r"{y'}",  txt)  # y-acute
+    txt = re.sub(u'ÿ|ÿ|ÿ',     r'{y"}',  txt)  # y-umlaut
+
+    txt = re.sub(u'Č|Č|Č',   r'{Cˇ}',  txt)  # C-caron
+    txt = re.sub(u'č|č|č',   r'{cˇ}',  txt)  # c-caron
+    txt = re.sub(u'Ď|Ď|Ď',   r'{Dˇ}',  txt)  # D-caron
+    txt = re.sub(u'ď|ď|ď',   r'{dˇ}',  txt)  # d-caron
+    txt = re.sub(u'Ě|Ě|Ě',   r'{Eˇ}',  txt)  # E-caron
+    txt = re.sub(u'ě|ě|ě',   r'{eˇ}',  txt)  # e-caron
+    txt = re.sub(u'Ĺ|Ĺ|Ĺ',   r"{L'}",  txt)  # L-acute
+    txt = re.sub(u'ĺ|ĺ|ĺ',   r"{l'}",  txt)  # l-acute
+    txt = re.sub(u'Ľ|Ľ|Ľ',   r'{Lˇ}',  txt)  # L-caron
+    txt = re.sub(u'ľ|ľ|ľ',   r'{lˇ}',  txt)  # l-caron
+    txt = re.sub(u'Ň|Ň|Ň',   r'{Nˇ}',  txt)  # N-caron
+    txt = re.sub(u'ň|ň|ň',   r'{nˇ}',  txt)  # n-caron
+
+    txt = re.sub(u'Œ|Œ|Œ',    r'{OE}',  txt)  # OE
+    txt = re.sub(u'œ|œ|œ',    r'{oe}',  txt)  # oe
+
+    txt = re.sub(u'Ŕ|Ŕ|Ŕ',   r"{R'}",  txt)  # R-acute
+    txt = re.sub(u'ŕ|ŕ|ŕ',   r"{r'}",  txt)  # r-acute
+    txt = re.sub(u'Ř|Ř|Ř',   r'{Rˇ}',  txt)  # R-caron
+    txt = re.sub(u'ř|ř|ř',   r'{rˇ}',  txt)  # r-caron
+    txt = re.sub(u'Ŝ|Ŝ',            r'{S^}',  txt)  # S-circumflex
+    txt = re.sub(u'ŝ|ŝ',            r'{s^}',  txt)  # s-circumflex
+    txt = re.sub(u'Š|Š|Š',   r'{Sˇ}',  txt)  # S-caron
+    txt = re.sub(u'š|š|š',   r'{sˇ}',  txt)  # s-caron
+    txt = re.sub(u'Ť|Ť|Ť',   r'{Tˇ}',  txt)  # T-caron
+    txt = re.sub(u'ť|ť|ť',   r'{tˇ}',  txt)  # t-caron
+    txt = re.sub(u'Ů|Ů|Ů',    r'{U°}',  txt)  # U-ring
+    txt = re.sub(u'ů|ů|ů',    r'{u°}',  txt)  # u-ring
+    txt = re.sub(u'Ž|Ž|Ž',   r'{Zˇ}',  txt)  # Z-caron
+    txt = re.sub(u'ž|ž|ž',   r'{zˇ}',  txt)  # z-caron
+
+    txt = re.sub(u'•|•|•',    r'{*}',   txt)  # bullet
+    txt = re.sub(u'₣|₣',           r'{Fr}',  txt)  # Franc
+    txt = re.sub(u'₤|₤',           r'{L=}',  txt)  # Lira
+    txt = re.sub(u'₨|₨',           r'{Rs}',  txt)  # Rupee
+    txt = re.sub(u'€|€|€',    r'{C=}',  txt)  # euro
+    txt = re.sub(u'™|™|™',   r'{tm}',  txt)  # trademark
+    txt = re.sub(u'♠|♠|♠',  r'{spade}',   txt)  # spade
+    txt = re.sub(u'♣|♣|♣',   r'{club}',    txt)  # club
+    txt = re.sub(u'♥|♥|♥',  r'{heart}',   txt)  # heart
+    txt = re.sub(u'♦|♦|♦',   r'{diamond}', txt)  # diamond
+
+    # Move into main code?
+    # txt = re.sub(u'\xa0',   r'p. ', txt)              # blank paragraph
+    # txt = re.sub(u'\n\n\n\n',   r'\n\np. \n\n', txt)  # blank paragraph
+    # txt = re.sub(u'\n  \n',   r'\n
    \n', txt)     # blank paragraph - br tag
+
+    return txt
diff --git a/ebook_converter/ebooks/txt/markdownml.py b/ebook_converter/ebooks/txt/markdownml.py
new file mode 100644
index 0000000..4c25c47
--- /dev/null
+++ b/ebook_converter/ebooks/txt/markdownml.py
@@ -0,0 +1,286 @@
+# -*- coding: utf-8 -*-
+from __future__ import absolute_import, division, print_function, unicode_literals
+
+__license__ = 'GPL 3'
+__copyright__ = '''2011, John Schember 
+2011, Leigh Parry '''
+__docformat__ = 'restructuredtext en'
+
+'''
+Transform OEB content into Textile formatted plain text
+'''
+import re
+
+from functools import partial
+
+from calibre.ebooks.htmlz.oeb2html import OEB2HTML
+from calibre.ebooks.oeb.base import XHTML, XHTML_NS, barename, namespace, rewrite_links
+from calibre.ebooks.oeb.stylizer import Stylizer
+from polyglot.builtins import unicode_type, string_or_bytes
+
+
+class MarkdownMLizer(OEB2HTML):
+
+    def extract_content(self, oeb_book, opts):
+        self.log.info('Converting XHTML to Markdown formatted TXT...')
+        self.opts = opts
+        self.in_code = False
+        self.in_pre = False
+        self.list = []
+        self.blockquotes = 0
+        self.remove_space_after_newline = False
+        self.base_hrefs = [item.href for item in oeb_book.spine]
+        self.map_resources(oeb_book)
+
+        self.style_bold = False
+        self.style_italic = False
+
+        txt = self.mlize_spine(oeb_book)
+
+        # Do some tidying up
+        txt = self.tidy_up(txt)
+
+        return txt
+
+    def mlize_spine(self, oeb_book):
+        output = ['']
+        for item in oeb_book.spine:
+            self.log.debug('Converting %s to Markdown formatted TXT...' % item.href)
+            self.rewrite_ids(item.data, item)
+            rewrite_links(item.data, partial(self.rewrite_link, page=item))
+            stylizer = Stylizer(item.data, item.href, oeb_book, self.opts, self.opts.output_profile)
+            output += self.dump_text(item.data.find(XHTML('body')), stylizer)
+            output.append('\n\n')
+        return ''.join(output)
+
+    def tidy_up(self, text):
+        # Remove blank space form beginning of paragraph.
+        text = re.sub('(?msu)^[ ]{1,3}', '', text)
+        # pre has 4 spaces. We trimmed 3 so anything with a space left is a pre.
+        text = re.sub('(?msu)^[ ]', '    ', text)
+
+        # Remove tabs that aren't at the beinning of a line
+        new_text = []
+        for l in text.splitlines():
+            start = re.match('\t+', l)
+            if start:
+                start = start.group()
+            else:
+                start = ''
+            l = re.sub('\t', '', l)
+            new_text.append(start + l)
+        text = '\n'.join(new_text)
+
+        # Remove spaces from blank lines.
+        text = re.sub('(?msu)^[ ]+$', '', text)
+
+        # Reduce blank lines
+        text = re.sub('(?msu)\n{7,}', '\n' * 6, text)
+
+        # Remove blank lines at beginning and end of document.
+        text = re.sub(r'^\s*', '', text)
+        text = re.sub(r'\s*$', '\n\n', text)
+
+        return text
+
+    def remove_newlines(self, text):
+        text = text.replace('\r\n', ' ')
+        text = text.replace('\n', ' ')
+        text = text.replace('\r', ' ')
+        # Condense redundant spaces created by replacing newlines with spaces.
+        text = re.sub(r'[ ]{2,}', ' ', text)
+        text = re.sub(r'\t+', '', text)
+        if self.remove_space_after_newline == True:  # noqa
+            text = re.sub(r'^ +', '', text)
+            self.remove_space_after_newline = False
+        return text
+
+    def prepare_string_for_markdown(self, txt):
+        txt = re.sub(r'([\\`*_{}\[\]()#+!])', r'\\\1', txt)
+        return txt
+
+    def prepare_string_for_pre(self, txt):
+        new_text = []
+        for l in txt.splitlines():
+            new_text.append('    ' + l)
+        return '\n'.join(new_text)
+
+    def dump_text(self, elem, stylizer):
+        '''
+        @elem: The element in the etree that we are working on.
+        @stylizer: The style information attached to the element.
+        '''
+
+        # We can only processes tags. If there isn't a tag return any text.
+        if not isinstance(elem.tag, string_or_bytes) \
+           or namespace(elem.tag) != XHTML_NS:
+            p = elem.getparent()
+            if p is not None and isinstance(p.tag, string_or_bytes) and namespace(p.tag) == XHTML_NS \
+                    and elem.tail:
+                return [elem.tail]
+            return ['']
+
+        # Setup our variables.
+        text = []
+        style = stylizer.style(elem)
+        tags = []
+        tag = barename(elem.tag)
+        attribs = elem.attrib
+
+        # Ignore anything that is set to not be displayed.
+        if style['display'] in ('none', 'oeb-page-head', 'oeb-page-foot') \
+           or style['visibility'] == 'hidden':
+            if hasattr(elem, 'tail') and elem.tail:
+                return [elem.tail]
+            return ['']
+
+        # Soft scene breaks.
+        if 'margin-top' in style.cssdict() and style['margin-top'] != 'auto':
+            ems = int(round(float(style.marginTop) / style.fontSize) - 1)
+            if ems >= 1:
+                text.append(u'\n\n' * ems)
+
+        bq = '> ' * self.blockquotes
+        # Block level elements
+        if tag in ('h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'p', 'div'):
+            h_tag = ''
+            if tag in ('h1', 'h2', 'h3', 'h4', 'h5', 'h6'):
+                h_tag = '#' * int(tag[1]) + ' '
+            text.append('\n' + bq + h_tag)
+            tags.append('\n')
+            self.remove_space_after_newline = True
+
+        if style['font-style'] == 'italic' or tag in ('i', 'em'):
+            if tag not in ('h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'cite'):
+                if self.style_italic == False:  # noqa
+                    text.append('*')
+                    tags.append('*')
+                    self.style_italic = True
+        if style['font-weight'] in ('bold', 'bolder') or tag in ('b', 'strong'):
+            if tag not in ('h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'th'):
+                if self.style_bold == False:  # noqa
+                    text.append('**')
+                    tags.append('**')
+                    self.style_bold = True
+        if tag == 'br':
+            text.append('  \n')
+            self.remove_space_after_newline = True
+        if tag == 'blockquote':
+            self.blockquotes += 1
+            tags.append('>')
+            text.append('> ' * self.blockquotes)
+        elif tag == 'code':
+            if not self.in_pre and not self.in_code:
+                text.append('`')
+                tags.append('`')
+                self.in_code = True
+        elif tag == 'pre':
+            if not self.in_pre:
+                text.append('\n')
+                tags.append('pre')
+                self.in_pre = True
+        elif tag == 'hr':
+            text.append('\n* * *')
+            tags.append('\n')
+        elif tag == 'a':
+            # Only write links with absolute (external) urls.
+            if self.opts.keep_links and 'href' in attribs and '://' in attribs['href']:
+                title = ''
+                if 'title' in attribs:
+                    title = ' "' + attribs['title'] + '"'
+                    remove_space = self.remove_space_after_newline
+                    title = self.remove_newlines(title)
+                    self.remove_space_after_newline = remove_space
+                text.append('[')
+                tags.append('](' + attribs['href'] + title + ')')
+        elif tag == 'img':
+            if self.opts.keep_image_references:
+                txt = '!'
+                if 'alt' in attribs:
+                    remove_space = self.remove_space_after_newline
+                    txt += '[' + self.remove_newlines(attribs['alt']) + ']'
+                    self.remove_space_after_newline = remove_space
+                txt += '(' + attribs['src'] + ')'
+                text.append(txt)
+        elif tag in ('ol', 'ul'):
+            tags.append(tag)
+            # Add the list to our lists of lists so we can track
+            # nested lists.
+            self.list.append({'name': tag, 'num': 0})
+        elif tag == 'li':
+            # Get the last list from our list of lists
+            if self.list:
+                li = self.list[-1]
+            else:
+                li = {'name': 'ul', 'num': 0}
+            # Add a new line to start the item
+            text.append('\n')
+            # Add indent if we have nested lists.
+            list_count = len(self.list)
+            # We only care about indenting nested lists.
+            if (list_count - 1) > 0:
+                text.append('\t' * (list_count - 1))
+            # Add blockquote if we have a blockquote in a list item.
+            text.append(bq)
+            # Write the proper sign for ordered and unorded lists.
+            if li['name'] == 'ul':
+                text.append('+ ')
+            elif li['name'] == 'ol':
+                li['num'] += 1
+                text.append(unicode_type(li['num']) + '. ')
+
+        # Process tags that contain text.
+        if hasattr(elem, 'text') and elem.text:
+            txt = elem.text
+            if self.in_pre:
+                txt = self.prepare_string_for_pre(txt)
+            elif self.in_code:
+                txt = self.remove_newlines(txt)
+            else:
+                txt = self.prepare_string_for_markdown(self.remove_newlines(txt))
+            text.append(txt)
+
+        # Recurse down into tags within the tag we are in.
+        for item in elem:
+            text += self.dump_text(item, stylizer)
+
+        # Close all open tags.
+        tags.reverse()
+        for t in tags:
+            if t in ('pre', 'ul', 'ol', '>'):
+                if t == 'pre':
+                    self.in_pre = False
+                    text.append('\n')
+                elif t == '>':
+                    self.blockquotes -= 1
+                elif t in ('ul', 'ol'):
+                    if self.list:
+                        self.list.pop()
+                    text.append('\n')
+            else:
+                if t == '**':
+                    self.style_bold = False
+                elif t == '*':
+                    self.style_italic = False
+                elif t == '`':
+                    self.in_code = False
+                text.append('%s' % t)
+
+        # Soft scene breaks.
+        if 'margin-bottom' in style.cssdict() and style['margin-bottom'] != 'auto':
+            ems = int(round((float(style.marginBottom) / style.fontSize) - 1))
+            if ems >= 1:
+                text.append(u'\n\n' * ems)
+
+        # Add the text that is outside of the tag.
+        if hasattr(elem, 'tail') and elem.tail:
+            tail = elem.tail
+            if self.in_pre:
+                tail = self.prepare_string_for_pre(tail)
+            elif self.in_code:
+                tail = self.remove_newlines(tail)
+            else:
+                tail = self.prepare_string_for_markdown(self.remove_newlines(tail))
+            text.append(tail)
+
+        return text
diff --git a/ebook_converter/ebooks/txt/newlines.py b/ebook_converter/ebooks/txt/newlines.py
new file mode 100644
index 0000000..56a890c
--- /dev/null
+++ b/ebook_converter/ebooks/txt/newlines.py
@@ -0,0 +1,32 @@
+# -*- coding: utf-8 -*-
+from __future__ import absolute_import, division, print_function, unicode_literals
+
+__license__ = 'GPL 3'
+__copyright__ = '2009, John Schember '
+__docformat__ = 'restructuredtext en'
+
+import os
+
+
+class TxtNewlines(object):
+
+    NEWLINE_TYPES = {
+                        'system'  : os.linesep,
+                        'unix'    : '\n',
+                        'old_mac' : '\r',
+                        'windows' : '\r\n'
+                     }
+
+    def __init__(self, newline_type):
+        self.newline = self.NEWLINE_TYPES.get(newline_type.lower(), os.linesep)
+
+
+def specified_newlines(newline, text):
+    # Convert all newlines to \n
+    text = text.replace('\r\n', '\n')
+    text = text.replace('\r', '\n')
+
+    if newline == '\n':
+        return text
+
+    return text.replace('\n', newline)
diff --git a/ebook_converter/ebooks/txt/textileml.py b/ebook_converter/ebooks/txt/textileml.py
new file mode 100644
index 0000000..e9eecfc
--- /dev/null
+++ b/ebook_converter/ebooks/txt/textileml.py
@@ -0,0 +1,502 @@
+# -*- coding: utf-8 -*-
+from __future__ import absolute_import, division, print_function, unicode_literals
+
+__license__ = 'GPL 3'
+__copyright__ = '2011, Leigh Parry '
+__docformat__ = 'restructuredtext en'
+
+'''
+Transform OEB content into Textile formatted plain text
+'''
+import re
+
+from functools import partial
+
+from calibre.ebooks.htmlz.oeb2html import OEB2HTML
+from calibre.ebooks.oeb.base import XHTML, XHTML_NS, barename, namespace, rewrite_links
+from calibre.ebooks.oeb.stylizer import Stylizer
+from calibre.ebooks import unit_convert
+from calibre.ebooks.textile.unsmarten import unsmarten
+from polyglot.builtins import string_or_bytes
+
+
+class TextileMLizer(OEB2HTML):
+
+    MAX_EM = 10
+
+    def extract_content(self, oeb_book, opts):
+        self.log.info('Converting XHTML to Textile formatted TXT...')
+        self.opts = opts
+        self.in_pre = False
+        self.in_table = False
+        self.links = {}
+        self.list = []
+        self.our_links = []
+        self.in_a_link = False
+        self.our_ids = []
+        self.images = {}
+        self.id_no_text = ''
+        self.style_embed = []
+        self.remove_space_after_newline = False
+        self.base_hrefs = [item.href for item in oeb_book.spine]
+        self.map_resources(oeb_book)
+
+        self.style_bold = False
+        self.style_italic = False
+        self.style_under = False
+        self.style_strike = False
+        self.style_smallcap = False
+
+        txt = self.mlize_spine(oeb_book)
+        if self.opts.unsmarten_punctuation:
+            txt = unsmarten(txt)
+
+        # Do some tidying up
+        txt = self.tidy_up(txt)
+
+        return txt
+
+    def mlize_spine(self, oeb_book):
+        output = ['']
+        for item in oeb_book.spine:
+            self.log.debug('Converting %s to Textile formatted TXT...' % item.href)
+            self.rewrite_ids(item.data, item)
+            rewrite_links(item.data, partial(self.rewrite_link, page=item))
+            stylizer = Stylizer(item.data, item.href, oeb_book, self.opts, self.opts.output_profile)
+            output += self.dump_text(item.data.find(XHTML('body')), stylizer)
+            output.append('\n\n')
+        return ''.join(output)
+
+    def tidy_up(self, text):
+        # May need tweaking and finetuning
+        def check_escaping(text, tests):
+            for t in tests:
+                # I'm not checking for duplicated spans '%' as any that follow each other were being incorrectly merged
+                txt = '%s' % t
+                if txt != '%':
+                    text = re.sub(r'([^'+t+'|^\n])'+t+r'\]\['+t+'([^'+t+'])', r'\1\2', text)
+                    text = re.sub(r'([^'+t+'|^\n])'+t+t+'([^'+t+'])', r'\1\2', text)
+                text = re.sub(r'(\s|[*_\'"])\[('+t+'[a-zA-Z0-9 \'",.*_]+'+t+r')\](\s|[*_\'"?!,.])', r'\1\2\3', text)
+            return text
+
+        # Now tidyup links and ids - remove ones that don't have a correponding opposite
+        if self.opts.keep_links:
+            for i in self.our_links:
+                if i[0] == '#':
+                    if i not in self.our_ids:
+                        text = re.sub(r'"(.+)":'+i+r'(\s)', r'\1\2', text)
+            for i in self.our_ids:
+                if i not in self.our_links:
+                    text = re.sub(r'%?\('+i+'\\)\xa0?%?', r'', text)
+
+        # Remove obvious non-needed escaping, add sub/sup-script ones
+        text = check_escaping(text, [r'\*', '_', r'\*'])
+        # escape the super/sub-scripts if needed
+        text = re.sub(r'(\w)([~^]\w+[~^])', r'\1[\2]', text)
+        # escape the super/sub-scripts if needed
+        text = re.sub(r'([~^]\w+[~^])(\w)', r'[\1]\2', text)
+
+        # remove empty spans
+        text = re.sub(r'%\xa0+', r'%', text)
+        # remove empty spans - MAY MERGE SOME ?
+        text = re.sub(r'%%', r'', text)
+        # remove spans from tagged output
+        text = re.sub(r'%([_+*-]+)%', r'\1', text)
+        # remove spaces before a newline
+        text = re.sub(r' +\n', r'\n', text)
+        # remove newlines at top of file
+        text = re.sub(r'^\n+', r'', text)
+        # correct blockcode paras
+        text = re.sub(r'\npre\.\n?\nbc\.', r'\nbc.', text)
+        # correct blockquote paras
+        text = re.sub(r'\nbq\.\n?\np.*?\. ', r'\nbq. ', text)
+
+        # reduce blank lines
+        text = re.sub(r'\n{3}', r'\n\np. \n\n', text)
+        text = re.sub(u'%\n(p[<>=]{1,2}\\.|p\\.)', r'%\n\n\1', text)
+        # Check span following blank para
+        text = re.sub(r'\n+ +%', r' %', text)
+        text = re.sub(u'p[<>=]{1,2}\\.\n\n?', r'', text)
+        # blank paragraph
+        text = re.sub(r'\n(p.*\.)\n', r'\n\1 \n\n', text)
+        # blank paragraph
+        text = re.sub(u'\n\xa0', r'\np. ', text)
+        # blank paragraph
+        text = re.sub(u'\np[<>=]{1,2}?\\. \xa0', r'\np. ', text)
+        text = re.sub(r'(^|\n)(p.*\. ?\n)(p.*\.)', r'\1\3', text)
+        text = re.sub(r'\n(p\. \n)(p.*\.|h.*\.)', r'\n\2', text)
+        # sort out spaces in tables
+        text = re.sub(r' {2,}\|', r' |', text)
+
+        # Now put back spaces removed earlier as they're needed here
+        text = re.sub(r'\np\.\n', r'\np. \n', text)
+        # reduce blank lines
+        text = re.sub(r' \n\n\n', r' \n\n', text)
+
+        return text
+
+    def remove_newlines(self, text):
+        text = text.replace('\r\n', ' ')
+        text = text.replace('\n', ' ')
+        text = text.replace('\r', ' ')
+        # Condense redundant spaces created by replacing newlines with spaces.
+        text = re.sub(r'[ ]{2,}', ' ', text)
+        text = re.sub(r'\t+', '', text)
+        if self.remove_space_after_newline == True:  # noqa
+            text = re.sub(r'^ +', '', text)
+            self.remove_space_after_newline = False
+        return text
+
+    def check_styles(self, style):
+        txt = '{'
+        if self.opts.keep_color:
+            if 'color' in style.cssdict() and style['color'] != 'black':
+                txt += 'color:'+style['color']+';'
+            if 'background' in style.cssdict():
+                txt += 'background:'+style['background']+';'
+        txt += '}'
+        if txt == '{}':
+            txt = ''
+        return txt
+
+    def check_halign(self, style):
+        tests = {'left':'<','justify':'<>','center':'=','right':'>'}
+        for i in tests:
+            if style['text-align'] == i:
+                return tests[i]
+        return ''
+
+    def check_valign(self, style):
+        tests = {'top':'^','bottom':'~'}  # , 'middle':'-'}
+        for i in tests:
+            if style['vertical-align'] == i:
+                return tests[i]
+        return ''
+
+    def check_padding(self, style, stylizer):
+        txt = ''
+        left_padding_pts = 0
+        left_margin_pts = 0
+        if 'padding-left' in style.cssdict() and style['padding-left'] != 'auto':
+            left_padding_pts = unit_convert(style['padding-left'], style.width, style.fontSize, stylizer.profile.dpi)
+        if 'margin-left' in style.cssdict() and style['margin-left'] != 'auto':
+            left_margin_pts = unit_convert(style['margin-left'], style.width, style.fontSize, stylizer.profile.dpi)
+        left = left_margin_pts + left_padding_pts
+        emleft = min(int(round(left / stylizer.profile.fbase)), self.MAX_EM)
+        if emleft >= 1:
+            txt += '(' * emleft
+        right_padding_pts = 0
+        right_margin_pts = 0
+        if 'padding-right' in style.cssdict() and style['padding-right'] != 'auto':
+            right_padding_pts = unit_convert(style['padding-right'], style.width, style.fontSize, stylizer.profile.dpi)
+        if 'margin-right' in style.cssdict() and style['margin-right'] != 'auto':
+            right_margin_pts = unit_convert(style['margin-right'], style.width, style.fontSize, stylizer.profile.dpi)
+        right = right_margin_pts + right_padding_pts
+        emright = min(int(round(right / stylizer.profile.fbase)), self.MAX_EM)
+        if emright >= 1:
+            txt += ')' * emright
+
+        return txt
+
+    def check_id_tag(self, attribs):
+        txt = ''
+        if 'id' in attribs:
+            txt = '(#'+attribs['id']+ ')'
+            self.our_ids.append('#'+attribs['id'])
+            self.id_no_text = u'\xa0'
+        return txt
+
+    def build_block(self, tag, style, attribs, stylizer):
+        txt = '\n' + tag
+        if self.opts.keep_links:
+            txt += self.check_id_tag(attribs)
+        txt += self.check_padding(style, stylizer)
+        txt += self.check_halign(style)
+        txt += self.check_styles(style)
+        return txt
+
+    def prepare_string_for_textile(self, txt):
+        if re.search(r'(\s([*&_+\-~@%|]|\?{2})\S)|(\S([*&_+\-~@%|]|\?{2})\s)', txt):
+            return ' ==%s== ' % txt
+        return txt
+
+    def dump_text(self, elem, stylizer):
+        '''
+        @elem: The element in the etree that we are working on.
+        @stylizer: The style information attached to the element.
+        '''
+
+        # We can only processes tags. If there isn't a tag return any text.
+        if not isinstance(elem.tag, string_or_bytes) \
+           or namespace(elem.tag) != XHTML_NS:
+            p = elem.getparent()
+            if p is not None and isinstance(p.tag, string_or_bytes) and namespace(p.tag) == XHTML_NS \
+                    and elem.tail:
+                return [elem.tail]
+            return ['']
+
+        # Setup our variables.
+        text = ['']
+        style = stylizer.style(elem)
+        tags = []
+        tag = barename(elem.tag)
+        attribs = elem.attrib
+
+        # Ignore anything that is set to not be displayed.
+        if style['display'] in ('none', 'oeb-page-head', 'oeb-page-foot') \
+           or style['visibility'] == 'hidden':
+            if hasattr(elem, 'tail') and elem.tail:
+                return [elem.tail]
+            return ['']
+
+        # Soft scene breaks.
+        if 'margin-top' in style.cssdict() and style['margin-top'] != 'auto':
+            ems = min(int(round(float(style.marginTop) / style.fontSize) - 1), self.MAX_EM)
+            if ems >= 1:
+                text.append(u'\n\n\xa0' * ems)
+
+        if tag in ('h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'p', 'div'):
+            if tag == 'div':
+                tag = 'p'
+            text.append(self.build_block(tag, style, attribs, stylizer))
+            text.append('. ')
+            tags.append('\n')
+
+        if style['font-style'] == 'italic' or tag in ('i', 'em'):
+            if tag not in ('h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'cite'):
+                if self.style_italic == False:  # noqa
+                    if self.in_a_link:
+                        text.append('_')
+                        tags.append('_')
+                    else:
+                        text.append('[_')
+                        tags.append('_]')
+                    self.style_embed.append('_')
+                    self.style_italic = True
+        if style['font-weight'] in ('bold', 'bolder') or tag in ('b', 'strong'):
+            if tag not in ('h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'th'):
+                if self.style_bold == False:  # noqa
+                    if self.in_a_link:
+                        text.append('*')
+                        tags.append('*')
+                    else:
+                        text.append('[*')
+                        tags.append('*]')
+                    self.style_embed.append('*')
+                    self.style_bold = True
+        if style['text-decoration'] == 'underline' or tag in ('u', 'ins'):
+            if tag != 'a':
+                if self.style_under == False:  # noqa
+                    text.append('[+')
+                    tags.append('+]')
+                    self.style_embed.append('+')
+                    self.style_under = True
+        if style['text-decoration'] == 'line-through' or tag in ('strike', 'del', 's'):
+            if self.style_strike == False:  # noqa
+                text.append('[-')
+                tags.append('-]')
+                self.style_embed.append('-')
+                self.style_strike = True
+        if tag == 'br':
+            for i in reversed(self.style_embed):
+                text.append(i)
+            text.append('\n')
+            for i in self.style_embed:
+                text.append(i)
+            tags.append('')
+            self.remove_space_after_newline = True
+        if tag == 'blockquote':
+            text.append('\nbq. ')
+            tags.append('\n')
+        elif tag in ('abbr', 'acronym'):
+            text.append('')
+            txt = attribs['title']
+            tags.append('(' + txt + ')')
+        elif tag == 'sup':
+            text.append('^')
+            tags.append('^')
+        elif tag == 'sub':
+            text.append('~')
+            tags.append('~')
+        elif tag == 'code':
+            if self.in_pre:
+                text.append('\nbc. ')
+                tags.append('')
+            else:
+                text.append('@')
+                tags.append('@')
+        elif tag == 'cite':
+            text.append('??')
+            tags.append('??')
+        elif tag == 'hr':
+            text.append('\n***')
+            tags.append('\n')
+        elif tag == 'pre':
+            self.in_pre = True
+            text.append('\npre. ')
+            tags.append('pre\n')
+        elif tag == 'a':
+            if self.opts.keep_links:
+                if 'href' in attribs:
+                    text.append('"')
+                    tags.append('a')
+                    tags.append('":' + attribs['href'])
+                    self.our_links.append(attribs['href'])
+                    if 'title' in attribs:
+                        tags.append('(' + attribs['title'] + ')')
+                    self.in_a_link = True
+                else:
+                    text.append('%')
+                    tags.append('%')
+        elif tag == 'img':
+            if self.opts.keep_image_references:
+                txt = '!' + self.check_halign(style)
+                txt += self.check_valign(style)
+                txt += attribs['src']
+                text.append(txt)
+                if 'alt' in attribs:
+                    txt = attribs['alt']
+                    if txt != '':
+                        text.append('(' + txt + ')')
+                tags.append('!')
+        elif tag in ('ol', 'ul'):
+            self.list.append({'name': tag, 'num': 0})
+            text.append('')
+            tags.append(tag)
+        elif tag == 'li':
+            if self.list:
+                li = self.list[-1]
+            else:
+                li = {'name': 'ul', 'num': 0}
+            text.append('\n')
+            if li['name'] == 'ul':
+                text.append('*' * len(self.list) + ' ')
+            elif li['name'] == 'ol':
+                text.append('#' * len(self.list) + ' ')
+            tags.append('')
+        elif tag == 'dl':
+            text.append('\n')
+            tags.append('')
+        elif tag == 'dt':
+            text.append('')
+            tags.append('\n')
+        elif tag == 'dd':
+            text.append('    ')
+            tags.append('')
+        elif tag == 'dd':
+            text.append('')
+            tags.append('\n')
+        elif tag == 'table':
+            txt = self.build_block(tag, style, attribs, stylizer)
+            txt += '. \n'
+            if txt != '\ntable. \n':
+                text.append(txt)
+            else:
+                text.append('\n')
+            tags.append('')
+        elif tag == 'tr':
+            txt = self.build_block('', style, attribs, stylizer)
+            txt += '. '
+            if txt != '\n. ':
+                txt = re.sub('\n', '', txt)
+                text.append(txt)
+            tags.append('|\n')
+        elif tag == 'td':
+            text.append('|')
+            txt = ''
+            txt += self.check_halign(style)
+            txt += self.check_valign(style)
+            if 'colspan' in attribs:
+                txt += '\\' + attribs['colspan']
+            if 'rowspan' in attribs:
+                txt += '/' + attribs['rowspan']
+            txt += self.check_styles(style)
+            if txt != '':
+                text.append(txt + '. ')
+            tags.append('')
+        elif tag == 'th':
+            text.append('|_. ')
+            tags.append('')
+        elif tag == 'span':
+            if style['font-variant'] == 'small-caps':
+                if self.style_smallcap == False:  # noqa
+                    text.append('&')
+                    tags.append('&')
+                    self.style_smallcap = True
+            else:
+                if self.in_a_link == False:  # noqa
+                    txt = '%'
+                    if self.opts.keep_links:
+                        txt += self.check_id_tag(attribs)
+                        txt += self.check_styles(style)
+                    if txt != '%':
+                        text.append(txt)
+                        tags.append('%')
+
+        if self.opts.keep_links and 'id' in attribs:
+            if tag not in ('body', 'div', 'h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'p', 'span', 'table'):
+                text.append(self.check_id_tag(attribs))
+
+        # Process the styles for any that we want to keep
+        if tag not in ('body', 'div', 'h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'p', 'hr', 'a', 'img',
+                'span', 'table', 'tr', 'td'):
+            if not self.in_a_link:
+                text.append(self.check_styles(style))
+
+        # Process tags that contain text.
+        if hasattr(elem, 'text') and elem.text:
+            txt = elem.text
+            if not self.in_pre:
+                txt = self.prepare_string_for_textile(self.remove_newlines(txt))
+            text.append(txt)
+            self.id_no_text = u''
+
+        # Recurse down into tags within the tag we are in.
+        for item in elem:
+            text += self.dump_text(item, stylizer)
+
+        # Close all open tags.
+        tags.reverse()
+        for t in tags:
+            if t in ('pre', 'ul', 'ol', 'li', 'table'):
+                if t == 'pre':
+                    self.in_pre = False
+                elif t in ('ul', 'ol'):
+                    if self.list:
+                        self.list.pop()
+                    if not self.list:
+                        text.append('\n')
+            else:
+                if t == 'a':
+                    self.in_a_link = False
+                    t = ''
+                text.append(self.id_no_text)
+                self.id_no_text = u''
+                if t in ('*]', '*'):
+                    self.style_bold = False
+                elif t in ('_]', '_'):
+                    self.style_italic = False
+                elif t == '+]':
+                    self.style_under = False
+                elif t == '-]':
+                    self.style_strike = False
+                elif t == '&':
+                    self.style_smallcap = False
+                if t in ('*]', '_]', '+]', '-]', '*', '_'):
+                    txt = self.style_embed.pop()
+                text.append('%s' % t)
+
+        # Soft scene breaks.
+        if 'margin-bottom' in style.cssdict() and style['margin-bottom'] != 'auto':
+            ems = min(int(round((float(style.marginBottom) / style.fontSize) - 1)), self.MAX_EM)
+            if ems >= 1:
+                text.append(u'\n\n\xa0' * ems)
+
+        # Add the text that is outside of the tag.
+        if hasattr(elem, 'tail') and elem.tail:
+            tail = elem.tail
+            if not self.in_pre:
+                tail = self.prepare_string_for_textile(self.remove_newlines(tail))
+            text.append(tail)
+
+        return text
diff --git a/ebook_converter/ebooks/txt/txtml.py b/ebook_converter/ebooks/txt/txtml.py
new file mode 100644
index 0000000..14cc7d8
--- /dev/null
+++ b/ebook_converter/ebooks/txt/txtml.py
@@ -0,0 +1,264 @@
+# -*- coding: utf-8 -*-
+from __future__ import absolute_import, division, print_function, unicode_literals
+
+__license__ = 'GPL 3'
+__copyright__ = '2009, John Schember '
+__docformat__ = 'restructuredtext en'
+
+'''
+Transform OEB content into plain text
+'''
+
+import re
+
+from lxml import etree
+from polyglot.builtins import string_or_bytes
+
+
+BLOCK_TAGS = [
+    'div',
+    'p',
+    'h1',
+    'h2',
+    'h3',
+    'h4',
+    'h5',
+    'h6',
+    'li',
+    'tr',
+]
+
+BLOCK_STYLES = [
+    'block',
+]
+
+HEADING_TAGS = [
+    'h1',
+    'h2',
+    'h3',
+    'h4',
+    'h5',
+    'h6',
+]
+
+SPACE_TAGS = [
+    'td',
+    'br',
+]
+
+
+class TXTMLizer(object):
+
+    def __init__(self, log):
+        self.log = log
+
+    def extract_content(self, oeb_book, opts):
+        self.log.info('Converting XHTML to TXT...')
+        self.oeb_book = oeb_book
+        self.opts = opts
+        self.toc_titles = []
+        self.toc_ids = []
+        self.last_was_heading = False
+
+        self.create_flat_toc(self.oeb_book.toc)
+
+        return self.mlize_spine()
+
+    def mlize_spine(self):
+        from calibre.ebooks.oeb.base import XHTML
+        from calibre.ebooks.oeb.stylizer import Stylizer
+        from calibre.utils.xml_parse import safe_xml_fromstring
+        output = [u'']
+        output.append(self.get_toc())
+        for item in self.oeb_book.spine:
+            self.log.debug('Converting %s to TXT...' % item.href)
+            for x in item.data.iterdescendants(etree.Comment):
+                if x.text and '--' in x.text:
+                    x.text = x.text.replace('--', '__')
+            content = etree.tostring(item.data, encoding='unicode')
+            content = self.remove_newlines(content)
+            content = safe_xml_fromstring(content)
+            stylizer = Stylizer(content, item.href, self.oeb_book, self.opts, self.opts.output_profile)
+            output += self.dump_text(content.find(XHTML('body')), stylizer, item)
+            output += '\n\n\n\n\n\n'
+        output = ''.join(output)
+        output = '\n'.join(l.rstrip() for l in output.splitlines())
+        output = self.cleanup_text(output)
+
+        return output
+
+    def remove_newlines(self, text):
+        self.log.debug('\tRemove newlines for processing...')
+        text = text.replace('\r\n', ' ')
+        text = text.replace('\n', ' ')
+        text = text.replace('\r', ' ')
+        # Condense redundant spaces created by replacing newlines with spaces.
+        text = re.sub(r'[ ]{2,}', ' ', text)
+
+        return text
+
+    def get_toc(self):
+        toc = ['']
+        if getattr(self.opts, 'inline_toc', None):
+            self.log.debug('Generating table of contents...')
+            toc.append('%s\n\n' % _('Table of Contents:'))
+            for item in self.toc_titles:
+                toc.append('* %s\n\n' % item)
+        return ''.join(toc)
+
+    def create_flat_toc(self, nodes):
+        '''
+        Turns a hierarchical list of TOC href's into a flat list.
+        '''
+        for item in nodes:
+            self.toc_titles.append(item.title)
+            self.toc_ids.append(item.href)
+            self.create_flat_toc(item.nodes)
+
+    def cleanup_text(self, text):
+        self.log.debug('\tClean up text...')
+        # Replace bad characters.
+        text = text.replace(u'\xa0', ' ')
+
+        # Replace tabs, vertical tags and form feeds with single space.
+        text = text.replace('\t+', ' ')
+        text = text.replace('\v+', ' ')
+        text = text.replace('\f+', ' ')
+
+        # Single line paragraph.
+        text = re.sub('(?<=.)\n(?=.)', ' ', text)
+
+        # Remove multiple spaces.
+        text = re.sub('[ ]{2,}', ' ', text)
+
+        # Remove excessive newlines.
+        text = re.sub('\n[ ]+\n', '\n\n', text)
+        if self.opts.remove_paragraph_spacing:
+            text = re.sub('\n{2,}', '\n', text)
+            text = re.sub(r'(?msu)^(?P[^\t\n]+?)$', lambda mo: u'%s\n\n' % mo.group('t'), text)
+            text = re.sub(r'(?msu)(?P[^\n])\n+(?P[^\t\n]+?)(?=\n)', lambda mo: '%s\n\n\n\n\n\n%s' % (mo.group('b'), mo.group('t')), text)
+        else:
+            text = re.sub('\n{7,}', '\n\n\n\n\n\n', text)
+
+        # Replace spaces at the beginning and end of lines
+        # We don't replace tabs because those are only added
+        # when remove paragraph spacing is enabled.
+        text = re.sub('(?imu)^[ ]+', '', text)
+        text = re.sub('(?imu)[ ]+$', '', text)
+
+        # Remove empty space and newlines at the beginning of the document.
+        text = re.sub(r'(?u)^[ \n]+', '', text)
+
+        if self.opts.max_line_length:
+            max_length = self.opts.max_line_length
+            if self.opts.max_line_length < 25 and not self.opts.force_max_line_length:
+                max_length = 25
+            short_lines = []
+            lines = text.splitlines()
+            for line in lines:
+                while len(line) > max_length:
+                    space = line.rfind(' ', 0, max_length)
+                    if space != -1:
+                        # Space was found.
+                        short_lines.append(line[:space])
+                        line = line[space + 1:]
+                    else:
+                        # Space was not found.
+                        if self.opts.force_max_line_length:
+                            # Force breaking at max_lenght.
+                            short_lines.append(line[:max_length])
+                            line = line[max_length:]
+                        else:
+                            # Look for the first space after max_length.
+                            space = line.find(' ', max_length, len(line))
+                            if space != -1:
+                                # Space was found.
+                                short_lines.append(line[:space])
+                                line = line[space + 1:]
+                            else:
+                                # No space was found cannot break line.
+                                short_lines.append(line)
+                                line = ''
+                # Add the text that was less than max_lengh to the list
+                short_lines.append(line)
+            text = '\n'.join(short_lines)
+
+        return text
+
+    def dump_text(self, elem, stylizer, page):
+        '''
+        @elem: The element in the etree that we are working on.
+        @stylizer: The style information attached to the element.
+        @page: OEB page used to determine absolute urls.
+        '''
+        from calibre.ebooks.oeb.base import XHTML_NS, barename, namespace
+
+        if not isinstance(elem.tag, string_or_bytes) \
+           or namespace(elem.tag) != XHTML_NS:
+            p = elem.getparent()
+            if p is not None and isinstance(p.tag, string_or_bytes) and namespace(p.tag) == XHTML_NS \
+                    and elem.tail:
+                return [elem.tail]
+            return ['']
+
+        text = ['']
+        style = stylizer.style(elem)
+
+        if style['display'] in ('none', 'oeb-page-head', 'oeb-page-foot') \
+           or style['visibility'] == 'hidden':
+            if hasattr(elem, 'tail') and elem.tail:
+                return [elem.tail]
+            return ['']
+
+        tag = barename(elem.tag)
+        tag_id = elem.attrib.get('id', None)
+        in_block = False
+        in_heading = False
+
+        # Are we in a heading?
+        # This can either be a heading tag or a TOC item.
+        if tag in HEADING_TAGS or '%s#%s' % (page.href, tag_id) in self.toc_ids:
+            in_heading = True
+            if not self.last_was_heading:
+                text.append('\n\n\n\n\n\n')
+
+        # Are we in a paragraph block?
+        if tag in BLOCK_TAGS or style['display'] in BLOCK_STYLES:
+            if self.opts.remove_paragraph_spacing and not in_heading:
+                text.append('\t')
+            in_block = True
+
+        if tag in SPACE_TAGS:
+            text.append(' ')
+
+        # Hard scene breaks.
+        if tag == 'hr':
+            text.append('\n\n* * *\n\n')
+        # Soft scene breaks.
+        try:
+            ems = int(round((float(style.marginTop) / style.fontSize) - 1))
+            if ems >= 1:
+                text.append('\n' * ems)
+        except:
+            pass
+
+        # Process tags that contain text.
+        if hasattr(elem, 'text') and elem.text:
+            text.append(elem.text)
+
+        # Recurse down into tags within the tag we are in.
+        for item in elem:
+            text += self.dump_text(item, stylizer, page)
+
+        if in_block:
+            text.append('\n\n')
+        if in_heading:
+            text.append('\n')
+            self.last_was_heading = True
+        else:
+            self.last_was_heading = False
+
+        if hasattr(elem, 'tail') and elem.tail:
+            text.append(elem.tail)
+
+        return text