diff --git a/src/calibre/ebooks/pml/pmlconverter.py b/src/calibre/ebooks/pml/pmlconverter.py
index b0d9ce1ec7..05cf488617 100644
--- a/src/calibre/ebooks/pml/pmlconverter.py
+++ b/src/calibre/ebooks/pml/pmlconverter.py
@@ -9,79 +9,390 @@ __copyright__ = '2009, John Schember <john@nachtimwald.com>'
 __docformat__ = 'restructuredtext en'
 
 import re
+import StringIO
 
 from calibre import my_unichr, prepare_string_for_xml
 from calibre.ebooks.pdb.ereader import image_name
 
-PML_HTML_RULES = [
-    # NOP-process all \x escapes, turning \\ into &#92;  This keeps the regex
-    # parsing simple while making sure that we don't try to honor \\x as \x
-    # (and also makes sure we DO honor \\\x as &#92; followed by \x).
-    (re.compile(r'\\(.)'), lambda match: '&#92;' if match.group(1) == '\\' else '\\' + match.group(1)),
+class PML_HTMLizer(object):
 
-    (re.compile(r'\\p'), lambda match: '<br /><br style="page-break-after: always;" />'),
-    (re.compile(r'\\x(?P<text>.*?)\\x', re.DOTALL), lambda match: '<h1 style="page-break-before: always;">%s</h1>' % match.group('text') if match.group('text') else ''),
-    (re.compile(r'\\X(?P<val>[0-4])(?P<text>.*?)\\X[0-4]', re.DOTALL), lambda match: '<h%s>%s</h%s>' % (int(match.group('val')) + 1, match.group('text'), int(match.group('val')) + 1) if match.group('text') else ''),
-    (re.compile(r'\\C\d=".+?"'), lambda match: ''), # This should be made to create a TOC entry
-    (re.compile(r'\\c(?P<text>.*?)\\c', re.DOTALL), lambda match: '<div style="text-align: center; margin: auto;">%s</div>' % match.group('text') if match.group('text') else ''),
-    (re.compile(r'\\r(?P<text>.*?)\\r', re.DOTALL), lambda match: '<div style="text-align: right;">%s</div>' % match.group('text') if match.group('text') else ''),
-    (re.compile(r'\\i(?P<text>.*?)\\i', re.DOTALL), lambda match: '<i>%s</i>' % match.group('text') if match.group('text') else ''),
-    (re.compile(r'\\u(?P<text>.*?)\\u', re.DOTALL), lambda match: '<u>%s</u>' % match.group('text') if match.group('text') else ''),
-    (re.compile(r'\\o(?P<text>.*?)\\o', re.DOTALL), lambda match: '<del>%s</del>' % match.group('text') if match.group('text') else ''),
-    (re.compile(r'\\v(?P<text>.*?)\\v', re.DOTALL), lambda match: '<!-- %s -->' % match.group('text') if match.group('text') else ''),
-    (re.compile(r'\\t(?P<text>.*?)\\t', re.DOTALL), lambda match: '<div style="margin-left: 5%%;">%s</div>' % match.group('text') if match.group('text') else ''),
-    (re.compile(r'\\T="(?P<val>\d+)%*"(?P<text>.*?)$', re.MULTILINE), lambda match: r'<div style="margin-left: %s%%">%s</div>' % (match.group('val'), match.group('text')) if match.group('text') else ''),
-    (re.compile(r'\\w="(?P<val>\d+)%"'), lambda match: '<hr width="%s%%" />' % match.group('val')),
-    (re.compile(r'\\n'), lambda match: ''),
-    (re.compile(r'\\s(?P<text>.*?)\\s', re.DOTALL), lambda match: '<span style="font-size: 1em;">%s</span>' % match.group('text') if match.group('text') else ''),
-    (re.compile(r'\\b(?P<text>.*?)\\b', re.DOTALL), lambda match: '<b>%s</b>' % match.group('text') if match.group('text') else ''), # \b is deprecated; \B should be used instead.
-    (re.compile(r'\\l(?P<text>.*?)\\l', re.DOTALL), lambda match: '<span style="font-size: 1.5em;">%s</span>' % match.group('text') if match.group('text') else ''),
-    (re.compile(r'\\B(?P<text>.*?)\\B', re.DOTALL), lambda match: '<b>%s</b>' % match.group('text') if match.group('text') else ''),
-    (re.compile(r'\\Sp(?P<text>.*?)\\Sp', re.DOTALL), lambda match: '<sup>%s</sup>' % match.group('text') if match.group('text') else ''),
-    (re.compile(r'\\Sb(?P<text>.*?)\\Sb', re.DOTALL), lambda match: '<sub>%s</sub>' % match.group('text') if match.group('text') else ''),
-    (re.compile(r'\\k(?P<text>.*?)\\k', re.DOTALL), lambda match: '<span style="font-size: 0.75em;">%s</span>' % match.group('text').upper() if match.group('text') else ''),
-    (re.compile(r'\\a(?P<num>\d{3})'), lambda match: '&#%s;' % match.group('num')),
-    (re.compile(r'\\U(?P<num>[0-9a-f]{4})'), lambda match: '%s' % my_unichr(int(match.group('num'), 16))),
-    (re.compile(r'\\m="(?P<name>.+?)"'), lambda match: '<img src="images/%s" />' % image_name(match.group('name')).strip('\x00')),
-    (re.compile(r'\\q="(?P<target>#.+?)"(?P<text>.*?)\\q', re.DOTALL), lambda match: '<a href="%s">%s</a>' % (match.group('target'), match.group('text')) if match.group('text') else ''),
-    (re.compile(r'\\Q="(?P<target>.+?)"'), lambda match: '<span id="%s"></span>' % match.group('target')),
-    (re.compile(r'\\-'), lambda match: ''),
-    (re.compile(r'\\Fn="(?P<target>.+?)"(?P<text>.*?)\\Fn'), lambda match: '<a href="#fns-%s">%s</a>' % (match.group('target'), match.group('text')) if match.group('text') else ''),
-    (re.compile(r'\\Sd="(?P<target>.+?)"(?P<text>.*?)\\Sd'), lambda match: '<a href="#fns-%s">%s</a>' % (match.group('target'), match.group('text')) if match.group('text') else ''),
-    # Just italicize index items as that is how the eReader software renders them.
-    (re.compile(r'\\I(?P<text>.*?)\\I', re.DOTALL), lambda match: '<i>%s</i>' % match.group('text') if match.group('text') else ''),
+    STATES = [
+        'i',
+        'u',
+        'd',
+        'b',
+        'sp',
+        'sb',
+        'h1',
+        'h2',
+        'h3',
+        'h4',
+        'h5',
+        'h6',
+        'a',
+        'c',
+        'r',
+        't',
+        's',
+        'l',
+        'T',
+        'Fn',
+        'Sd',
+        'FS'
+    ]
 
-    # Sidebar and Footnotes
-    (re.compile(r'&lt;sidebar\s+id="(?P<target>.+?)"&gt;\s*(?P<text>.*?)\s*&lt;/sidebar&gt;', re.DOTALL), lambda match: '<div id="fns-%s">%s</div>' % (match.group('target'), match.group('text')) if match.group('text') else ''),
-    (re.compile(r'&lt;footnote\s+id="(?P<target>.+?)"&gt;\s*(?P<text>.*?)\s*&lt;/footnote&gt;', re.DOTALL), lambda match: '<div id="fns-%s">%s</div>' % (match.group('target'), match.group('text')) if match.group('text') else ''),
+    STATES_VALUE_REQ = [
+        'a',
+        'T',
+        'FS'
+    ]
 
-    # eReader files are one paragraph per line.
-    # This forces the lines to wrap properly.
-    (re.compile('^(?P<text>.+)$', re.MULTILINE), lambda match: '<p>%s</p>' % match.group('text')),
-    # Remove empty <p>'s.
-    (re.compile('<p>[ ]*</p>'), lambda match: ''),
-    # Ensure empty lines carry over.
-    (re.compile('(\r\n|\n|\r){3}'), lambda match: '<br />'),
+    STATES_TAGS = {
+        'h1': ('<h1 style="page-break-after: always;">', '</h1>'),
+        'h2': ('<h2>', '</h2>'),
+        'h3': ('<h3>', '</h3>'),
+        'h4': ('<h4>', '</h4>'),
+        'h5': ('<h5>', '</h5>'),
+        'h6': ('<h6>', '</h6>'),
+        'sp': ('<sup>', '</sup>'),
+        'sb': ('<sub>', '</sub>'),
+        'a': ('<a href="%s">', '</a>'),
+        'c': ('<div style="text-align: center; margin: auto;">', '</div>'),
+        'r': ('<div style="text-align: right;">', '</div>'),
+        't': ('<div style="margin-left: 5%;">', '</div>'),
+        'T': ('<div style="margin-left: %s;">', '</div>'),
+        'i': ('<span style="font-style : italic;">', '</span>'),
+        'u': ('<span style="text-decoration : underline;">', '</span>'),
+        'd': ('<span style="text-decoration: line-through;">', '</span>'),
+        'b': ('<span style="font-weight: bold;">', '</span>'),
+        's': ('<span style="font-size: 75%">', '</span>'),
+        'l': ('<span style="font-size: 125%">', '</span>'),
+        'FS': ('<div id="%s">', '</div>'),
+    }
 
-    # Try to fix some of the misordering of character-attribute tags.
-    (re.compile(r'(?P<ch>(<(i|u|b|del|sup|sub)( [^>]+)?>)+)(?P<close>(</(div|span)>)+)'), lambda match: match.group('close') + match.group('ch')),
-    (re.compile(r'(?P<ch>(<(i|u|b|del|sup|sub|span)( [^>]+)?>)+)(?P<blk>(<(div|h\d)( [^>]+)?>)+)'), lambda match: match.group('blk') + match.group('ch')),
+    CODE_STATES = {
+        'q': 'a',
+        'x': 'h1',
+        'X0': 'h2',
+        'X1': 'h3',
+        'X2': 'h4',
+        'X3': 'h5',
+        'X4': 'h6',
+        'Sp': 'sp',
+        'Sb': 'sb',
+        'c': 'c',
+        'r': 'r',
+        't': 't',
+        'T': 'T',
+        'i': 'i',
+        'I': 'i',
+        'u': 'u',
+        'o': 'd',
+        'b': 'b',
+        'B': 'b',
+        's': 's',
+        'l': 'l',
+        'Fn': 'a',
+        'Sd': 'a',
+        'FN': 'FS',
+        'SB': 'FS',
+    }
 
-    # Remove unmatched plm codes.
-    (re.compile(r'\\X[0-4]'), lambda match: ''),
-    (re.compile(r'\\T="\d+%*"'), lambda match: ''),
-    (re.compile(r'\\Sp'), lambda match: ''),
-    (re.compile(r'\\Sb'), lambda match: ''),
-    # Remove invalid single item pml codes.
-    (re.compile(r'\\.'), lambda match: ''),
-]
+    DIV_STATES = [
+        'c',
+        'r',
+        't',
+        'T',
+        'FS',
+    ]
 
-def pml_to_html(pml):
-    html = prepare_string_for_xml(pml)
-    for rule in PML_HTML_RULES:
-        html = rule[0].sub(rule[1], html)
+    SPAN_STATES = [
+        's',
+        'l',
+        'i',
+        'u',
+        'd',
+        'b',
+    ]
 
-    return html
+    def __init__(self, close_all):
+        self.close_all = close_all
+        self.state = {}
+
+    def prepare_pml(self, pml):
+        # Remove comments
+        pml = re.sub(r'(?mus)\\v(?P<text>.*?)\\v', '', pml)
+        # Footnotes and Sidebars
+        pml = re.sub(r'(?mus)<footnote\s+id="(?P<target>.+?)">\s*(?P<text>.*?)\s*</footnote>', lambda match: '\\FN="fns-%s"%s\\FN' % (match.group('target'), match.group('text')) if match.group('text') else '', pml)
+        pml = re.sub(r'(?mus)<sidebar\s+id="(?P<target>.+?)">\s*(?P<text>.*?)\s*</sidebar>', lambda match: '\\SB="fns-%s"%s\\SB' % (match.group('target'), match.group('text')) if match.group('text') else '', pml)
+
+        pml = prepare_string_for_xml(pml)
+
+        pml = re.sub(r'\\a(?P<num>\d{3})', lambda match: '&#%s;' % match.group('num'), pml)
+        pml = re.sub(r'\\U(?P<num>[0-9a-f]{4})', lambda match: '%s' % my_unichr(int(match.group('num'), 16)), pml)
+
+        return pml
+
+    def prepare_line(self, line):
+        line = re.sub(r'[ ]{2,}', ' ', line)
+        line = re.sub(r'^[ ]*(?=.)', '', line)
+        line = re.sub(r'(?<=.)[ ]*$', '', line)
+        line = re.sub(r'^[ ]*$', '', line)
+
+        return line
+
+    def start_line(self):
+        start = u''
+
+        for key, val in self.state.items():
+            if val[0]:
+                if key not in self.STATES_VALUE_REQ:
+                    start += self.STATES_TAGS[key][0]
+                else:
+                    start += self.STATES_TAGS[key][0] % val[1]
+
+        return u'<p>%s' % start
+
+    def end_line(self):
+        end = u''
+
+        for key, val in self.state.items():
+            if val[0]:
+                if key == 'T':
+                    self.state['T'][0] = False
+                end += self.STATES_TAGS[key][1]
+
+        return u'%s</p>' % end
+
+    def process_code_simple(self, code):
+        if code not in self.CODE_STATES.keys():
+            return u''
+
+        text = u''
+
+        if self.state[self.CODE_STATES[code]][0]:
+            text = self.STATES_TAGS[self.CODE_STATES[code]][1]
+        else:
+            text = self.STATES_TAGS[self.CODE_STATES[code]][0]
+
+        self.state[self.CODE_STATES[code]][0] = not self.state[self.CODE_STATES[code]][0]
+
+        return text
+
+    def process_code_link(self, stream, pre=''):
+        text = u''
+
+        href = self.code_value(stream)
+        if pre:
+            href = '#%s-%s' % (pre, href)
+
+        if self.state['a'][0]:
+            text = self.STATES_TAGS['a'][1]
+        else:
+            text = self.STATES_TAGS['a'][0] % href
+            self.state['a'][1] = href
+
+        self.state['a'][0] = not self.state['a'][0]
+
+        return text
+
+    def process_code_div_span(self, code, stream):
+        if self.close_all:
+            return self.process_code_div_span_call(code, stream)
+        else:
+            return self.process_code_div_span_ind(code, stream)
+
+    def process_code_div_span_ind(self, code, stream):
+        text = u''
+        ds = []
+
+        code = self.CODE_STATES[code]
+
+        if code in self.DIV_STATES:
+            ds = self.DIV_STATES[:]
+        elif code in self.SPAN_STATES:
+            ds = self.SPAN_STATES[:]
+
+        if self.state[code][0]:
+            # Close all.
+            for c in ds:
+                if self.state[c][0]:
+                    text += self.STATES_TAGS[c][1]
+            # Reopen the based on state.
+            del ds[ds.index(code)]
+            for c in ds:
+                if self.state[c][0]:
+                    if c in self.STATES_VALUE_REQ:
+                        text += self.STATES_TAGS[self.CODE_STATES[c]][0] % self.state[c][1]
+                    else:
+                        text += self.STATES_TAGS[c][0]
+        else:
+            if code in self.STATES_VALUE_REQ:
+                val = self.code_value(stream)
+                text = self.STATES_TAGS[code][0] % val
+                self.state[code][1] = val
+            else:
+                text = self.STATES_TAGS[code][0]
+
+        self.state[code][0] = not self.state[code][0]
+
+        return text
+
+    def process_code_div_span_call(self, code, stream):
+        text = u''
+        divs = self.DIV_STATES[:]
+        spans = self.SPAN_STATES[:]
+
+        code = self.CODE_STATES[code]
+
+        if self.state[code][0]:
+            # Close all divs then spans.
+            for c in spans+divs:
+                if self.state[c][0]:
+                    text += self.STATES_TAGS[c][1]
+            # Reopen the based on state. Open divs then spans
+            if code in self.DIV_STATES:
+                del divs[divs.index(code)]
+            if code in self.SPAN_STATES:
+                del spans[spans.index(code)]
+            for c in divs+spans:
+                if state[c][0]:
+                    if c in self.STATES_VALUE_REQ:
+                        text += self.STATES_TAGS[self.CODE_STATES[c]][0] % self.state[c][1]
+                    else:
+                        text += self.STATES_TAGS[c][0]
+        else:
+            if code in self.STATES_VALUE_REQ:
+                val = self.code_value(stream)
+                text = self.STATES_TAGS[code][0] % val
+                state[code][1] = val
+            else:
+                text = self.STATES_TAGS[code][0]
+
+        self.state[code][0] = not self.state[code][0]
+
+        return text
+
+    def code_value(self, stream):
+        value = u''
+        open = False
+
+        c = stream.read(1)
+        while c != '':
+            if open and c != '"':
+                value += c
+            if c == '"':
+                if not open:
+                    open = True
+                else:
+                    break
+            c = stream.read(1)
+
+        return value.strip()
+
+    def parse_pml(self, pml):
+        pml = self.prepare_pml(pml)
+        output = []
+
+        self.state = {}
+        for s in self.STATES:
+            self.state[s] = [False, ''];
+
+        for line in pml.splitlines():
+            if not line:
+                continue
+            parsed = []
+            empty = True
+
+            # Must use StringIO, cStringIO does not support unicode
+            line = StringIO.StringIO(self.prepare_line(line))
+            parsed.append(self.start_line())
+
+            c = line.read(1)
+            while c != '':
+                text = u''
+
+                if c == '\\':
+                    c = line.read(1)
+
+                    if c == 'x':
+                        text = self.process_code_simple(c)
+                    elif c in 'XS':
+                        l = line.read(1)
+                        if '%s%s' % (c, l) == 'Sd':
+                            text = self.process_code_link(line, 'fns')
+                        elif '%s%s' % (c, l) == 'SB':
+                            text = self.process_code_div_span('SB', line)
+                        else:
+                            text = self.process_code_simple('%s%s' % (c, l))
+                    elif c == 'q':
+                        text = self.process_code_link(line)
+                    elif c in 'crtTiIuobB':
+                        text = self.process_code_div_span(c, line)
+                    elif c in 'sl':
+                        close = u''
+                        if c == 's' and self.state['l']:
+                            close = self.process_code_div_span('l', line)
+                        if c == 'l' and self.state['s']:
+                            close = self.process_code_div_span('s', line)
+                        text = self.process_code_div_span(c, line)
+                        text = close+text
+                    elif c == 'm':
+                        empty = False
+                        src = self.code_value(line)
+                        text = '<img src="images/%s" />' % image_name(src).strip('\x00')
+                    elif c == 'Q':
+                        empty = False
+                        id = self.code_value(line)
+                        text = '<span id="%s"></span>' % id
+                    elif c == 'p':
+                        empty = False
+                        text = '<br /><br style="page-break-after: always;" />'
+                    elif c == 'C':
+                        # This should be made to create a TOC entry
+                        line.read(1)
+                        self.code_value(line)
+                    elif c == 'n':
+                        pass
+                    elif c == 'F':
+                        l = line.read(1)
+                        if '%s%s' % (c, l) == 'Fn':
+                            text = self.process_code_link(line, 'fns')
+                        elif '%s%s' % (c, l) == 'FN':
+                            text = self.process_code_div_span('FN', line)
+                    elif c == 'w':
+                        empty = False
+                        text = '<hr width="%s" />' % self.code_value(line)
+                    elif c == '-':
+                        empty = False
+                        text = '&shy;'
+                    elif c == '\\':
+                        empty = False
+                        text = '\\'
+                else:
+                    if c != ' ':
+                        empty = False
+                    text = c
+                parsed.append(text)
+                c = line.read(1)
+
+            if not empty:
+                text = self.end_line()
+                parsed.append(text)
+                output.append(u''.join(parsed))
+            line.close()
+
+        return u'\n'.join(output)
+
+
+def pml_to_html(pml, close_all=False):
+    '''
+    close_all will close div all div and span tags when one is closed and then
+    re-open the appropriate ones.
+    '''
+
+    hizer = PML_HTMLizer(close_all)
+    return hizer.parse_pml(pml)
 
 def footnote_sidebar_to_html(id, pml):
     if id.startswith('\x01'):