# -*- coding: utf-8 -*-

import sys, re, string, urllib, htmlentitydefs
import edbm


def wrap(text, width):
    """
    A word-wrap function that preserves existing line breaks
    and most spaces in the text. Expects that existing line
    breaks are posix newlines (\n).
    """
    return reduce(lambda line, word, width=width: '%s%s%s' % (line, [' ', '\n    '][(len(line[line.rfind('\n')+1:]) + len(word.split('\n',1)[0]) >= width)], word), text.split(' '))

def url_escape(text):
    return urllib.quote_plus(text)

def html_escape(text):
    text = string.replace(text, '&', '&')
    text = string.replace(text, '<', '&lt;')
    text = string.replace(text, '>', '&gt;')
    return text
                          
def iso2ascii(text):
    for c in '\xc0\xc1\xc2\xc3\xc5\xe0\xe1\xe2\xe3\xe5':
        text = string.replace(text, c, 'a')
    for c in '\xc7\xe7':
        text = string.replace(text, c, 'c')
    for c in '\xc8\xc9\xca\xcb\xe8\xe9\xea\xeb':
        text = string.replace(text, c, 'e')
    for c in '\xcc\xcd\xce\xcf\xec\xed\xee\xef':
        text = string.replace(text, c, 'i')
    for c in '\xd1\xf1':
        text = string.replace(text, c, 'n')
    for c in '\xd2\xd3\xd4\xd5\xf2\xf3\xf4\xf5':
        text = string.replace(text, c, 'o')
    for c in '\xd9\xda\xdb\xf9\xfa\xfb':
        text = string.replace(text, c, 'u')
    for c in '\xdd\xfd\xff':
        text = string.replace(text, c, 'y')
    text = string.replace(text, '\xe4', 'ae')
    text = string.replace(text, '\xf6', 'oe')
    text = string.replace(text, '\xfc', 'ue')
    text = string.replace(text, '\xc4', 'Ae')
    text = string.replace(text, '\xd6', 'Oe')
    text = string.replace(text, '\xdc', 'Ue')
    text = string.replace(text, '\xdf', 'ss')
    text = string.replace(text, '\xd8', 'o')
    text = string.replace(text, '\xf8', 'O')
    text = string.replace(text, '\xc6', 'ae')
    text = string.replace(text, '\xe6', 'AE')
    return text

def utf82ascii(text):
    chars = {
        'A'  : ['\xc3\x80', '\xc3\x81', '\xc3\x82', '\xc3\x83', '\xc3\x85', '\xc4\x80', '\xc4\x82', '\xc4\x84'],
        'C'  : ['\xc3\x87', '\xc4\x86', '\xc4\x88', '\xc4\x8a', '\xc4\x8c'],
        'D'  : ['\xc4\x8e', '\xc4\x90'],
        'E'  : ['\xc3\x88', '\xc3\x89', '\xc3\x8a', '\xc3\x8b', '\xc4\x92', '\xc4\x94', '\xc4\x96', '\xc4\x98', '\xc4\x9a'],
        'G'  : ['\xc4\x9c', '\xc4\x9e', '\xc4\xa0', '\xc4\xa2'],
        'H'  : ['\xc4\xa4', '\xc4\xa6'],
        'I'  : ['\xc3\x8c', '\xc3\x8d', '\xc3\x8e', '\xc3\x8f', '\xc4\xa8', '\xc4\xaa', '\xc4\xac', '\xc4\xae', '\xc4\xb0'],
        'J'  : ['\xc4\xb4'],
        'K'  : ['\xc4\xb6'],
        'L'  : ['\xc4\xb9', '\xc4\xbb', '\xc4\xbd', '\xc4\xbf', '\xc5\x81'],
        'N'  : ['\xc3\x91', '\xc5\x83', '\xc5\x85', '\xc5\x87'],
        'O'  : ['\xc3\x92', '\xc3\x93', '\xc3\x94', '\xc3\x95', '\xc3\x96', '\xc3\x98', '\xc5\x8c', '\xc5\x8e', '\xc5\x90'],
        'R'  : ['\xc5\x94', '\xc5\x96', '\xc5\x98'],
        'S'  : ['\xc5\x9a', '\xc5\x9c', '\xc5\x9e', '\xc5\xa0'],
        'T'  : ['\xc5\xa2', '\xc5\xa4', '\xc5\xa6'],
        'U'  : ['\xc3\x99', '\xc3\x9a', '\xc3\x9b', '\xc5\xa8', '\xc5\xaa', '\xc5\xac', '\xc5\xae', '\xc5\xb0', '\xc5\xb2'],
        'W'  : ['\xc5\xb4'],
        'Y'  : ['\xc3\x9d', '\xc5\xb6', '\xc5\xb8'],
        'Z'  : ['\xc5\xb9', '\xc5\xbb', '\xc5\xbd'],
        'a'  : ['\xc3\xa0', '\xc3\xa1', '\xc3\xa2', '\xc3\xa3', '\xc3\xa5', '\xc4\x81', '\xc4\x83', '\xc4\x85'],
        'c'  : ['\xc3\xa7', '\xc4\x87', '\xc4\x89', '\xc4\x8b', '\xc4\x8d'],
        'd'  : ['\xc4\x8f', '\xc4\x91'],
        'e'  : ['\xc3\xa8', '\xc3\xa9', '\xc3\xaa', '\xc3\xab', '\xc4\x93', '\xc4\x95', '\xc4\x97', '\xc4\x99', '\xc4\x9b'],
        'g'  : ['\xc4\x9d', '\xc4\x9f', '\xc4\xa1', '\xc4\xa3'],
        'h'  : ['\xc4\xa5', '\xc4\xa7'],
        'i'  : ['\xc3\xac', '\xc3\xad', '\xc3\xae', '\xc3\xaf', '\xc4\xa9', '\xc4\xab', '\xc4\xad', '\xc4\xaf', '\xc4\xb1'],
        'j'  : ['\xc4\xb5'],
        'k'  : ['\xc4\xb7'],
        'l'  : ['\xc4\xba', '\xc4\xbc', '\xc4\xbe', '\xc5\x80', '\xc5\x82'],
        'n'  : ['\xc3\xb1', '\xc5\x84', '\xc5\x86', '\xc5\x88', '\xc5\x89'],
        'o'  : ['\xc3\xb2', '\xc3\xb3', '\xc3\xb4', '\xc3\xb5', '\xc3\xb8', '\xc5\x8d', '\xc5\x8f', '\xc5\x91'],
        'r'  : ['\xc5\x95', '\xc5\x97', '\xc5\x99'],
        's'  : ['\xc5\x9b', '\xc5\x9d', '\xc5\x9f', '\xc5\xa1'],
        't'  : ['\xc5\xa3', '\xc5\xa5', '\xc5\xa7'],
        'u'  : ['\xc3\xb9', '\xc3\xba', '\xc3\xbb', '\xc5\xa9', '\xc5\xab', '\xc5\xad', '\xc5\xaf', '\xc5\xb1', '\xc5\xb3'],
        'w'  : ['\xc5\xb5'],
        'y'  : ['\xc3\xbd', '\xc3\xbf', '\xc5\xb7'],
        'z'  : ['\xc5\xba', '\xc5\xbc', '\xc5\xbe'],
    }

    umlauts = {
        'Ae' : ['\xc3\x84'],
        'Oe' : ['\xc3\x96'],
        'Ue' : ['\xc3\x9c'],
        'ae' : ['\xc3\xa4', '\xc3\xa6'],
        'oe' : ['\xc3\xb6', '\xc5\x93'],
        'ue' : ['\xc3\xbc'],
        'ss' : ['\xc3\x9f'],
        'AE' : ['\xc3\x86'],
        'OE' : ['\xc5\x92'],
    }

    greek = {
        r'\Alpha '     : ['\xce\x91'],
        r'\Beta '      : ['\xce\x92'],
        r'\Gamma '     : ['\xce\x93'],
        r'\Delta '     : ['\xce\x94'],
        r'\Epsilon '   : ['\xce\x95'],
        r'\Zeta '      : ['\xce\x96'],
        r'\Eta '       : ['\xce\x97'],
        r'\Theta '     : ['\xce\x98'],
        r'\Iota '      : ['\xce\x99'],
        r'\Kappa '     : ['\xce\x9a'],
        r'\Lamda '     : ['\xce\x9b'],
        r'\Mu '        : ['\xce\x9c'],
        r'\Nu '        : ['\xce\x9d'],
        r'\Xi '        : ['\xce\x9e'],
        r'\Omicron '   : ['\xce\x9f'],
        r'\Pi '        : ['\xce\xa0'],
        r'\Rho '       : ['\xce\xa1'],
        r'\Sigma '     : ['\xce\xa3'],
        r'\Tau '       : ['\xce\xa4'],
        r'\Upsilon '   : ['\xce\xa5'],
        r'\Phi '       : ['\xce\xa6'],
        r'\Chi '       : ['\xce\xa7'],
        r'\Psi '       : ['\xce\xa8'],
        r'\Omega '     : ['\xce\xa9'],
        r'\alpha '     : ['\xce\xb1'],
        r'\beta '      : ['\xce\xb2'],
        r'\gamma '     : ['\xce\xb3'],
        r'\delta '     : ['\xce\xb4'],
        r'\epsilon'    : ['\xcf\xb5'],
        r'\zeta '      : ['\xce\xb6'],
        r'\eta '       : ['\xce\xb7'],
        r'\theta'      : ['\xce\xb8'],
        r'\iota '      : ['\xce\xb9'],
        r'\kappa '     : ['\xce\xba'],
        r'\lamda '     : ['\xce\xbb'],
        r'\mu '        : ['\xce\xbc'],
        r'\nu '        : ['\xce\xbd'],
        r'\xi '        : ['\xce\xbe'],
        r'\omicron '   : ['\xce\xbf'],
        r'\pi'         : ['\xcf\x80'],
        r'\rho'        : ['\xcf\x81'],
        r'\sigma'      : ['\xcf\x83'],
        r'\tau '       : ['\xcf\x84'],
        r'\upsilon '   : ['\xcf\x85'],
        r'\phi'        : ['\xcf\x95'],
        r'\chi '       : ['\xcf\x87'],
        r'\psi '       : ['\xcf\x88'],
        r'\omega '     : ['\xcf\x89'],
        r'\varepsilon' : ['\xce\xb5'],
        r'\vartheta'   : ['\xcf\x91'],
        r'\varpi'      : ['\xcf\x96'],
        r'\varrho'     : ['\xcf\xb1'],
        r'\varsigma'   : ['\xcf\x82'],
        r'\varphi'     : ['\xcf\x86'],
    }

    for char, list in umlauts.items() + chars.items():
        for code in list:
            text = string.replace(text, code, char)
    return text


def text2html(text):
    for key, value in htmlentitydefs.codepoint2name.items():
        if key < 256 and key != 34:
            text = text.replace(chr(key), "&%s;" % (value))
    text = re.sub('[\x80-\xff]+', ' ', text)
    return text


_ascii_patterns = [
    [r"\\\" *([AOUaou])","\\1e", 0],
    [r"\\[\"H] *{([AOUaou])}","\\1e", 0],
    [r"\\H +([AOUaou])","\\1e", 0],
    [r"{\\[`'^~=.\"uvtcdbHBD] *{(.)}}", "\\1", 0],
    [r"{\\[`'^~=.\"uvtcdbHBD] *(.)}", "\\1", 0],
    [r"\\[`'^~=.\"uvtcdbHBD] *{(.)}", "\\1", 0],
    [r"\\[`'^~=.\"] *", "", 0],
    [r"\\[uvtcdbHBD] +", "", 0],
    [r"(\\[`'^~=.\"uvtcdbHBD])[\\{]", "", 1],
    [r"{\\['\"`^~=.] *\\i}", "i", 0],
    [r"{\\['\"`^~=.uvtcdbHBD] *{\\i}}", "i", 0],
    [r"\\([ijlL]) *{(.?)}", "\\1\\2", 0],
    [r"\\([ijlL]) +", "\\1", 0],
    [r"(\\i)([\\{\d\W]|$)", "i", 1],
    [r"(\\j)([\\{\d\W]|$)", "j", 1],
    [r"{\\([ijlL]) *}", "\\1", 0],
    [r"(\\l)([\\{\d\W]|$)", "l", 1],
    [r"(\\L)([\\{\d\W]|$)", "L", 1],
    [r"\\AA *{(.?)}","A\\1", 0],
    [r"(\\AA)([\\{\d\W]|$)", "A", 1],
    [r"\\AA +","A", 0],
    [r"{\\AA *}","A", 0],
    [r"\\AE *{(.?)}","AE\\1", 0],
    [r"\\AE +","AE", 0],
    [r"(\\AE)([\\{\d\W]|$)","AE", 1],
    [r"{\\AE *}","AE", 0],
    [r"\\O *{(.?)}","OE\\1", 0],
    [r"(\\O)([\\{\d\W]|$)","OE", 1],
    [r"\\O +","OE", 0],
    [r"{\\O *}","OE", 0],
    [r"\\o *{(.?)}","oe\\1", 0],
    [r"(\\o)([\\{\d\W]|$)","oe", 1],
    [r"\\o +","oe", 0],
    [r"{\\o *}","oe", 0],
    [r"\\OE *{(.?)}", "OE\\1", 0],
    [r"\\OE +","OE", 0],
    [r"(\\OE)([\\{\d\W]|$)", "OE", 1],
    [r"{\\OE *}","OE", 0],
    [r"\\oe *{(.?)}", "oe\\1", 0],
    [r"\\oe +","oe", 0],
    [r"(\\oe)([\\{\d\W]|$)", "oe", 1],
    [r"{\\oe *}","oe", 0],
    [r"\\aa *{(.?)}","a\\1", 0],
    [r"\\aa +","a", 0],
    [r"(\\aa)([\\{\d\W]|$)", "a", 1],
    [r"{\\aa *}","a", 0],
    [r"\\ae *{(.?)}","ae\\1", 0],
    [r"\\ae +","ae", 0],
    [r"(\\ae)([\\{\d\W]|$)", "ae", 1],
    [r"{\\ae *}","ae", 0],
    [r"\\uu *{(.?)}", "u\\1", 0],
    [r"\\uu +","u", 0],
    [r"(\\uu)([\\{\d\W]|$)", "u", 1],
    [r"{\\uu *}","u", 0],
    [r"\\oo *{(.?)}", "o\\1", 0],
    [r"\\oo +","o", 0],
    [r"(\\oo)([\\{\d\W]|$)", "o", 1],
    [r"{\\oo *}", "o", 0],
    [r"\\UU *{(.?)}", "U\\1", 0],
    [r"\\UU +","U", 0],
    [r"(\\UU)([\\{\d\W]|$)", "U", 1],
    [r"{\\UU *}","U", 0],
    [r"\\[^\"`'] *\\j( *{}| +|$)", "j", 0],
    [r"\\[^\"`'] *\\j([\d\W])", "j\\1", 0],
    [r"\\dj *{(.?)}", "dj\\1", 0],
    [r"\\dj +", "dj", 0],
    [r"(\\dj)([\\{\d\W]|$)", "d", 1],
    [r"{\\dj *}", "dj", 0],
    [r"\\Dj *{(.?)}", "Dj\\1", 0],
    [r"\\Dj +", "Dj", 0],
    [r"(\\Dj)(\\| *{)", "Dj", 0],
    [r"{\\Dj *}", "Dj", 0],
    [r"\\oe *{(.?)}", "oe\\1", 0],
    [r"\\oe +","oe", 0],
    [r"(\\oe)([\\{\d\W]|$)","oe", 1],
    [r"\\ss *{(.?)}","ss\\1", 0],
    [r"\\ss +","ss", 0],
    [r"(\\ss)([\\{\d\W]|$)", "ss", 1],
    [r"{\\ss *}","ss", 0],
]
edbm.chconv.set_ascii_patterns(_ascii_patterns)
                      
def tex2ascii(text):
    text = edbm.chconv.tex2ascii(text)
    text = re.sub("{([a-zA-Z]+)}", r"\1", text)
    return text


_iso_patterns = [
    [r'\\[H"] *{(\w)}', 5, 0],
    [r'\\" *(\w)',5, 0],
    [r'{\\" *(\w)}',5, 0],
    [r'{\\" *{\\(.)}}',5, 0],
    [r"\\` *{(\w)}",2, 0],
    [r"\\` *(\w)",2, 0],
    [r"{\\` *(\w)}",2, 0],
    [r"\\' *{(\w)}",1, 0],
    [r"\\' *(\w)",1, 0],
    [r"{\\' *(\w)}",1, 0],
    [r"\\\^ *{(\w)}",3, 0],
    [r"\\\^ *(\w)",3, 0],
    [r"{\\\^ *(\w)}",3, 0],
    [r"\\~ *{(\w)}",4, 0],
    [r"\\~ *(\w)",4, 0],
    [r"{\\~ *(\w)}",4, 0],
    [r"\\H +(\w)",5, 0],
    [r"{\\H +(\w)}",5, 0],
    [r"\\' *{\\(.)}", 1, 0],
    [r"\\' *\\(.)( +|$)", 1, 0],
    [r"{\\' *\\(.)}", 1, 0],
    [r"{\\' *{\\(.)}}", 1, 0],
    [r"\\` *{\\(.)}", 2, 0],
    [r"\\` *\\(.)( +|$)", 2, 0],
    [r"{\\` *\\(.)}", 2, 0],
    [r"\\\^ *{\\(.)}", 3, 0],
    [r"\\\^ *\\(.)( +|$)", 3, 0],
    [r"{\\\^ *\\(.)}", 3, 0],
    [r"{\\\^ *{\\(.)}}", 3, 0],
    [r'{\\["H] *\\i}', "\357", 0],
    [r'\\["H] *\\i *{(.?)}',"\357\\1", 0],
    [r'\\["H] *\\i( +|$)',"\357", 0],
    [r'(\\["H] *\\i)([\d\W]|$)',"\357", 1],
    [r"\\' *\\i *{(.?)}","\355\\1", 0],
    [r"(\\' *\\i)([\d\W]|$)","\355", 1],
    [r'\\["H] *{\\i *}',"\357", 0],
    [r"\\[^\"`'] *\\j( *{}| +|$)", "j", 0],
    [r"\\[^\"`'] *\\j([\d\W])", "j\\1", 0],
    [r"\\c( +c| *{c})","\347", 0],
    [r"{\\c{c}}","\347", 0],
    [r"\\c( +C| *{C})","\307", 0],
    [r"\\[`'^~=.\"uvtcdbHBD] *{(.)}", "\\1", 0],
    [r"{\\[`'^~=.\"uvtcdbHBD] *{(.)}}", "\\1", 0],
    [r"\\[`'^~=.\"] *", "", 0],
    [r"{\\[`'^~=.\"]([a-zA-Z])}", "\\1", 0],
    [r"\\[uvtcdbHBD] +", "", 0],
    [r"{\\[uvtcdbHBD]{\\(.)}}", "\\1", 0],
    [r"(\\[`'^~=.\"uvtcdbHBD])[\\{]", "", 1],
    [r"\\([ijlL]) *{(.?)}", "\\1\\2", 0],
    [r"\\([ijlL]) +", "\\1", 0],
    [r"(\\i)([\d\W]|$)", "i", 1],
    [r"(\\j)([\d\W]|$)", "j", 1],
    [r"{\\([ijlL]) *}", "\\1", 0],
    [r"(\\l)([\d\W]|$)", "l", 1],
    [r"(\\L)([\d\W]|$)", "L", 1],
    [r"\\AA *{(.?)}","\305\\1", 0],
    [r"(\\AA)([\d\W]|$)", "\305", 0],
    [r"\\AA +","\305", 0],
    [r"{\\AA *}","\305", 0],
    [r"\\AE *{(.?)}","AE\\1", 0],
    [r"\\AE +","AE", 0],
    [r"(\\AE)([\d\W]|$)","AE", 1],
    [r"{\\AE *}","AE", 0],
    [r"\\O *{(.?)}","\330\\1", 0],
    [r"(\\O)([\d\W]|$)","\330", 1],
    [r"\\O +","\330", 0],
    [r"{\\O *}","\330", 0],
    [r"\\o *{(.?)}","\370\\1", 0],
    [r"(\\o)([\d\W]|$)","\370", 1],
    [r"\\o +","\370", 0],
    [r"{\\o *}","\370", 0],
    [r"\\OE *{(.?)}", "OE\\1", 0],
    [r"\\OE +","OE", 0],
    [r"(\\OE)([\d\W]|$)", "OE", 1],
    [r"{\\OE *}","OE", 0],
    [r"\\aa *{(.?)}","\345\\1", 0],
    [r"\\aa +","\345", 0],
    [r"(\\aa)([\d\W]|$)", "\345", 1],
    [r"{\\aa *}","\345", 0],
    [r"\\ae *{(.?)}","ae\\1", 0],
    [r"\\ae +","ae", 0],
    [r"(\\ae)([\d\W]|$)", "ae", 1],
    [r"{\\ae *}","ae", 0],
    [r"\\uu *{(.?)}", "u\\1", 0],
    [r"\\uu +","u", 0],
    [r"(\\uu)([\d\W]|$)", "u", 1],
    [r"{\\uu *}","u", 0],
    [r"\\oo *{(.?)}", "o\\1", 0],
    [r"\\oo +","o", 0],
    [r"(\\oo)([\d\W]|$)", "o", 1],
    [r"{\\oo *}", "o", 0],
    [r"\\UU *{(.?)}", "U\\1", 0],
    [r"\\UU +","U", 0],
    [r"(\\UU)([\d\W]|$)", "U", 1],
    [r"{\\UU *}","U", 0],
    [r"\\[^\"`'] *\\j( *{}| +|$)", "j", 0],
    [r"\\[^\"`'] *\\j([\d\W])", "j\\1", 0],
    [r"\\dj *{(.?)}", "dj\\1", 0],
    [r"\\dj +", "dj", 0],
    [r"(\\dj)([\d\W]|$)", "dj", 1],
    [r"{\\dj *}", "dj", 0],
    [r"\\Dj *{(.?)}", "Dj\\1", 0],
    [r"\\Dj +", "Dj", 0],
    [r"(\\Dj)([\d\W]|$)", "Dj", 0],
    [r"{\\Dj *}", "Dj", 0],
    [r"\\oe *{(.?)}", "oe\\1", 0],
    [r"\\oe +","oe", 0],
    [r"(\\oe)([\d\W]|$)","oe", 1],
    [r"{\\oe *}","oe", 0],
    [r"\\ss *{(.?)}","\337\\1", 0],
    [r"\\ss +","\337", 0],
    [r"(\\ss)([\d\W]|$)", "\337", 1],
    [r"{\\ss *}","\337", 0],
]

edbm.chconv.set_iso_patterns(_iso_patterns)

def tex2iso(text):
    text =  edbm.chconv.tex2iso(text)
    text = re.sub("{([a-zA-Z]+)}", r"\1", text)
    return text


_utf8_patterns = [
    [r"\\' *(\w)", 1, 0],
    [r"\\' *{(\w)}", 1, 0],
    [r"{\\' *(\w)}", 1, 0],
    [r"{\\' *{\\([ijIJ])}}", 1, 0],
    [r"\\` *(\w)", 2, 0],
    [r"\\` *{(\w)}", 2, 0],
    [r"{\\` *(\w)}", 2, 0],
    [r"\\\^ *(\w)", 3, 0],
    [r"\\\^ *{(\w)}", 3, 0],
    [r"{\\\^ *(\w)}", 3, 0],
    [r"\\~ *(\w)", 4, 0],
    [r"\\~ *{(\w)}", 4, 0],
    [r"{\\~ *(\w)}", 4, 0],
    [r'\\\"(\w)', 5, 0],
    [r'\\\"{(\w)}', 5, 0],
    [r'{\\\"(\w)}', 5, 0],
    [r'\\H +(\w)', 5, 0],
    [r'\\H +{(\w)}', 5, 0],
    [r'{\\H +(\w)}', 5, 0],
    [r"\\c +(\w)", 6, 0],
    [r"\\c *{(\w)}", 6, 0],
    [r"{\\c +(\w)}", 6, 0],
    [r"\\=(\w)", 7, 0],
    [r"\\={(\w)}", 7, 0],
    [r"{\\=(\w)}", 7, 0],
    [r"\\B +(\w)", 7, 0],
    [r"\\B +{(\w)}", 7, 0],
    [r"{\\B +(\w)}", 7, 0],
    [r"\\u +(\w)", 8, 0],
    [r"\\u *{(\w)}", 8, 0],
    [r"{\\u +(\w)}", 8, 0],
    [r"\\\.(\w)", 9, 0],
    [r"\\\.{(\w)}", 9, 0],
    [r"{\\\.(\w)}", 9, 0],
    [r"\\D +(\w)", 9, 0],
    [r"\\D +{(\w)}", 9, 0],
    [r"{\\D +(\w)}", 9, 0],
    [r"\\v +(\w)", 10, 0],
    [r"\\v *{(\w)}", 10, 0],
    [r"{\\v +(\w)}", 10, 0],
    [r"\\d +(\w)", 11, 0],
    [r"\\d *{(\w)}", 11, 0],
    [r"{\\d +(\w)}", 11, 0],
    [r"\\b +(\w)", 12, 0],
    [r"\\b *{(\w)}", 12, 0],
    [r"{\\b +(\w)}", 12, 0],
    [r"\\' *\\AE( +|$)", 0, 0x01fc],
    [r"\\AA( +|$)", 0, 0x00c5],
    [r"{\\AA}", 0, 0x00c5],
    [r"\\aa( +|$)", 0, 0x00e5],
    [r"{\\aa}", 0, 0x00e5],
    [r"{\\l}", 0, 0x0142],
    [r"{\\L}", 0, 0x0141],
    [r"\\l( +|$)", 0, 0x0142],
    [r"\\L( +|$)", 0, 0x0141],
    [r"\\i( +|$)", 0, 0x69],
    [r"{\\i}", 0, 0x69],
    [r"{\\o}", 0, 0x00f8],
    [r"\\o +", 0, 0x00f8],
    [r"{\\O}", 0, 0x00d8],
    [r"\\O +", 0, 0x00d8],
]

edbm.chconv.set_utf8_patterns(_utf8_patterns)

def tex2utf8(text):
    text = re.sub(r'{\\([ij])}', r' \1', text)
    text = string.replace(text, r'{\ss}', 'ss')
    text = string.replace(text, r'{\ae}', 'ae')
    text = string.replace(text, r'{\oe}', 'oe')
    text = string.replace(text, r'{\uu}', 'uu')
    text = edbm.chconv.tex2utf8(text)

    if 0:
        text = string.replace(text, r'$\Alpha$', '\xce\x91')
        text = string.replace(text, r'$\Beta$', '\xce\x92')
        text = string.replace(text, r'$\Gamma$', '\xce\x93')
        text = string.replace(text, r'$\Delta$', '\xce\x94')
        text = string.replace(text, r'$\Epsilon$', '\xce\x95')
        text = string.replace(text, r'$\Zeta$', '\xce\x96')
        text = string.replace(text, r'$\Eta$', '\xce\x97')
        text = string.replace(text, r'$\Theta$', '\xce\x98')
        text = string.replace(text, r'$\Iota$', '\xce\x99')
        text = string.replace(text, r'$\Kappa$', '\xce\x9a')
        text = string.replace(text, r'$\Lamda$', '\xce\x9b')
        text = string.replace(text, r'$\Mu$', '\xce\x9c')
        text = string.replace(text, r'$\Nu$', '\xce\x9d')
        text = string.replace(text, r'$\Xi$', '\xce\x9e')
        text = string.replace(text, r'$\Omicron$', '\xce\x9f')
        text = string.replace(text, r'$\Pi$', '\xce\xa0')
        text = string.replace(text, r'$\Rho$', '\xce\xa1')
        text = string.replace(text, r'$\Sigma$', '\xce\xa3')
        text = string.replace(text, r'$\Tau$', '\xce\xa4')
        text = string.replace(text, r'$\Upsilon$', '\xce\xa5')
        text = string.replace(text, r'$\Phi$', '\xce\xa6')
        text = string.replace(text, r'$\Chi$', '\xce\xa7')
        text = string.replace(text, r'$\Psi$', '\xce\xa8')
        text = string.replace(text, r'$\Omega$', '\xce\xa9')
        text = string.replace(text, r'$\alpha$', '\xce\xb1')
        text = string.replace(text, r'$\beta$', '\xce\xb2')
        text = string.replace(text, r'$\gamma$', '\xce\xb3')
        text = string.replace(text, r'$\delta$', '\xce\xb4')
        text = string.replace(text, r'$\epsilon$', '\xcf\xb5')
        text = string.replace(text, r'$\zeta$', '\xce\xb6')
        text = string.replace(text, r'$\eta$', '\xce\xb7')
        text = string.replace(text, r'$\theta$', '\xce\xb8')
        text = string.replace(text, r'$\iota$', '\xce\xb9')
        text = string.replace(text, r'$\kappa$', '\xce\xba')
        text = string.replace(text, r'$\lamda$', '\xce\xbb')
        text = string.replace(text, r'$\mu$', '\xce\xbc')
        text = string.replace(text, r'$\nu$', '\xce\xbd')
        text = string.replace(text, r'$\xi$', '\xce\xbe')
        text = string.replace(text, r'$\omicron$', '\xce\xbf')
        text = string.replace(text, r'$\pi$', '\xcf\x80')
        text = string.replace(text, r'$\rho$', '\xcf\x81')
        text = string.replace(text, r'$\sigma$', '\xcf\x83')
        text = string.replace(text, r'$\tau$', '\xcf\x84')
        text = string.replace(text, r'$\upsilon$', '\xcf\x85')
        text = string.replace(text, r'$\phi$', '\xcf\x95')
        text = string.replace(text, r'$\chi$', '\xcf\x87')
        text = string.replace(text, r'$\psi$', '\xcf\x88')
        text = string.replace(text, r'$\omega$', '\xcf\x89')
        text = string.replace(text, r'$\varepsilon$', '\xce\xb5')
        text = string.replace(text, r'$\vartheta$', '\xcf\x91')
        text = string.replace(text, r'$\varpi$', '\xcf\x96')
        text = string.replace(text, r'$\varrho$', '\xcf\xb1')
        text = string.replace(text, r'$\varsigma$', '\xcf\x82')
        text = string.replace(text, r'$\varphi$', '\xcf\x86')
        text = string.replace(text, r'\Alpha', '\xce\x91')
        text = string.replace(text, r'\Beta', '\xce\x92')
        text = string.replace(text, r'\Gamma', '\xce\x93')
        text = string.replace(text, r'\Delta', '\xce\x94')
        text = string.replace(text, r'\Epsilon', '\xce\x95')
        text = string.replace(text, r'\Zeta', '\xce\x96')
        text = string.replace(text, r'\Eta', '\xce\x97')
        text = string.replace(text, r'\Theta', '\xce\x98')
        text = string.replace(text, r'\Iota', '\xce\x99')
        text = string.replace(text, r'\Kappa', '\xce\x9a')
        text = string.replace(text, r'\Lamda', '\xce\x9b')
        text = string.replace(text, r'\Mu', '\xce\x9c')
        text = string.replace(text, r'\Nu', '\xce\x9d')
        text = string.replace(text, r'\Xi', '\xce\x9e')
        text = string.replace(text, r'\Omicron', '\xce\x9f')
        text = string.replace(text, r'\Pi', '\xce\xa0')
        text = string.replace(text, r'\Rho', '\xce\xa1')
        text = string.replace(text, r'\Sigma', '\xce\xa3')
        text = string.replace(text, r'\Tau', '\xce\xa4')
        text = string.replace(text, r'\Upsilon', '\xce\xa5')
        text = string.replace(text, r'\Phi', '\xce\xa6')
        text = string.replace(text, r'\Chi', '\xce\xa7')
        text = string.replace(text, r'\Psi', '\xce\xa8')
        text = string.replace(text, r'\Omega', '\xce\xa9')
        text = string.replace(text, r'\alpha', '\xce\xb1')
        text = string.replace(text, r'\beta', '\xce\xb2')
        text = string.replace(text, r'\gamma', '\xce\xb3')
        text = string.replace(text, r'\delta', '\xce\xb4')
        text = string.replace(text, r'\epsilon', '\xcf\xb5')
        text = string.replace(text, r'\zeta', '\xce\xb6')
        text = string.replace(text, r'\eta', '\xce\xb7')
        text = string.replace(text, r'\theta', '\xce\xb8')
        text = string.replace(text, r'\iota', '\xce\xb9')
        text = string.replace(text, r'\kappa', '\xce\xba')
        text = string.replace(text, r'\lamda', '\xce\xbb')
        text = string.replace(text, r'\mu', '\xce\xbc')
        text = string.replace(text, r'\nu', '\xce\xbd')
        text = string.replace(text, r'\xi', '\xce\xbe')
        text = string.replace(text, r'\omicron', '\xce\xbf')
        text = string.replace(text, r'\pi', '\xcf\x80')
        text = string.replace(text, r'\rho', '\xcf\x81')
        text = string.replace(text, r'\sigma', '\xcf\x83')
        text = string.replace(text, r'\tau', '\xcf\x84')
        text = string.replace(text, r'\upsilon', '\xcf\x85')
        text = string.replace(text, r'\phi', '\xcf\x95')
        text = string.replace(text, r'\chi', '\xcf\x87')
        text = string.replace(text, r'\psi', '\xcf\x88')
        text = string.replace(text, r'\omega', '\xcf\x89')
        text = string.replace(text, r'\varepsilon', '\xce\xb5')
        text = string.replace(text, r'\vartheta', '\xcf\x91')
        text = string.replace(text, r'varpi', '\xcf\x96')
        text = string.replace(text, r'\varrho', '\xcf\xb1')
        text = string.replace(text, r'\varsigma', '\xcf\x82')
        text = string.replace(text, r'\varphi', '\xcf\x86')
    return text