htmlentitydefs.py

来自「mallet是自然语言处理、机器学习领域的一个开源项目。」· Python 代码 · 共 258 行 · 第 1/2 页

"""HTML character entity references."""

entitydefs = {
    'AElig':    '\306',         # latin capital letter AE = latin capital ligature AE, U+00C6 ISOlat1
    'Aacute':   '\301',         # latin capital letter A with acute, U+00C1 ISOlat1
    'Acirc':    '\302',         # latin capital letter A with circumflex, U+00C2 ISOlat1
    'Agrave':   '\300',         # latin capital letter A with grave = latin capital letter A grave, U+00C0 ISOlat1
    'Alpha':    'Α',       # greek capital letter alpha, U+0391
    'Aring':    '\305',         # latin capital letter A with ring above = latin capital letter A ring, U+00C5 ISOlat1
    'Atilde':   '\303',         # latin capital letter A with tilde, U+00C3 ISOlat1
    'Auml':     '\304',         # latin capital letter A with diaeresis, U+00C4 ISOlat1
    'Beta':     'Β',       # greek capital letter beta, U+0392
    'Ccedil':   '\307',         # latin capital letter C with cedilla, U+00C7 ISOlat1
    'Chi':      'Χ',       # greek capital letter chi, U+03A7
    'Dagger':   '‡',      # double dagger, U+2021 ISOpub
    'Delta':    'Δ',       # greek capital letter delta, U+0394 ISOgrk3
    'ETH':      '\320',         # latin capital letter ETH, U+00D0 ISOlat1
    'Eacute':   '\311',         # latin capital letter E with acute, U+00C9 ISOlat1
    'Ecirc':    '\312',         # latin capital letter E with circumflex, U+00CA ISOlat1
    'Egrave':   '\310',         # latin capital letter E with grave, U+00C8 ISOlat1
    'Epsilon':  'Ε',       # greek capital letter epsilon, U+0395
    'Eta':      'Η',       # greek capital letter eta, U+0397
    'Euml':     '\313',         # latin capital letter E with diaeresis, U+00CB ISOlat1
    'Gamma':    'Γ',       # greek capital letter gamma, U+0393 ISOgrk3
    'Iacute':   '\315',         # latin capital letter I with acute, U+00CD ISOlat1
    'Icirc':    '\316',         # latin capital letter I with circumflex, U+00CE ISOlat1
    'Igrave':   '\314',         # latin capital letter I with grave, U+00CC ISOlat1
    'Iota':     'Ι',       # greek capital letter iota, U+0399
    'Iuml':     '\317',         # latin capital letter I with diaeresis, U+00CF ISOlat1
    'Kappa':    'Κ',       # greek capital letter kappa, U+039A
    'Lambda':   'Λ',       # greek capital letter lambda, U+039B ISOgrk3
    'Mu':       'Μ',       # greek capital letter mu, U+039C
    'Ntilde':   '\321',         # latin capital letter N with tilde, U+00D1 ISOlat1
    'Nu':       'Ν',       # greek capital letter nu, U+039D
    'OElig':    'Œ',       # latin capital ligature OE, U+0152 ISOlat2
    'Oacute':   '\323',         # latin capital letter O with acute, U+00D3 ISOlat1
    'Ocirc':    '\324',         # latin capital letter O with circumflex, U+00D4 ISOlat1
    'Ograve':   '\322',         # latin capital letter O with grave, U+00D2 ISOlat1
    'Omega':    'Ω',       # greek capital letter omega, U+03A9 ISOgrk3
    'Omicron':  'Ο',       # greek capital letter omicron, U+039F
    'Oslash':   '\330',         # latin capital letter O with stroke = latin capital letter O slash, U+00D8 ISOlat1
    'Otilde':   '\325',         # latin capital letter O with tilde, U+00D5 ISOlat1
    'Ouml':     '\326',         # latin capital letter O with diaeresis, U+00D6 ISOlat1
    'Phi':      'Φ',       # greek capital letter phi, U+03A6 ISOgrk3
    'Pi':       'Π',       # greek capital letter pi, U+03A0 ISOgrk3
    'Prime':    '″',      # double prime = seconds = inches, U+2033 ISOtech
    'Psi':      'Ψ',       # greek capital letter psi, U+03A8 ISOgrk3
    'Rho':      'Ρ',       # greek capital letter rho, U+03A1
    'Scaron':   'Š',       # latin capital letter S with caron, U+0160 ISOlat2
    'Sigma':    'Σ',       # greek capital letter sigma, U+03A3 ISOgrk3
    'THORN':    '\336',         # latin capital letter THORN, U+00DE ISOlat1
    'Tau':      'Τ',       # greek capital letter tau, U+03A4
    'Theta':    'Θ',       # greek capital letter theta, U+0398 ISOgrk3
    'Uacute':   '\332',         # latin capital letter U with acute, U+00DA ISOlat1
    'Ucirc':    '\333',         # latin capital letter U with circumflex, U+00DB ISOlat1
    'Ugrave':   '\331',         # latin capital letter U with grave, U+00D9 ISOlat1
    'Upsilon':  'Υ',       # greek capital letter upsilon, U+03A5 ISOgrk3
    'Uuml':     '\334',         # latin capital letter U with diaeresis, U+00DC ISOlat1
    'Xi':       'Ξ',       # greek capital letter xi, U+039E ISOgrk3
    'Yacute':   '\335',         # latin capital letter Y with acute, U+00DD ISOlat1
    'Yuml':     'Ÿ',       # latin capital letter Y with diaeresis, U+0178 ISOlat2
    'Zeta':     'Ζ',       # greek capital letter zeta, U+0396
    'aacute':   '\341',         # latin small letter a with acute, U+00E1 ISOlat1
    'acirc':    '\342',         # latin small letter a with circumflex, U+00E2 ISOlat1
    'acute':    '\264',         # acute accent = spacing acute, U+00B4 ISOdia
    'aelig':    '\346',         # latin small letter ae = latin small ligature ae, U+00E6 ISOlat1
    'agrave':   '\340',         # latin small letter a with grave = latin small letter a grave, U+00E0 ISOlat1
    'alefsym':  'ℵ',      # alef symbol = first transfinite cardinal, U+2135 NEW
    'alpha':    'α',       # greek small letter alpha, U+03B1 ISOgrk3
    'amp':      '\46',          # ampersand, U+0026 ISOnum
    'and':      '∧',      # logical and = wedge, U+2227 ISOtech
    'ang':      '∠',      # angle, U+2220 ISOamso
    'aring':    '\345',         # latin small letter a with ring above = latin small letter a ring, U+00E5 ISOlat1
    'asymp':    '≈',      # almost equal to = asymptotic to, U+2248 ISOamsr
    'atilde':   '\343',         # latin small letter a with tilde, U+00E3 ISOlat1
    'auml':     '\344',         # latin small letter a with diaeresis, U+00E4 ISOlat1
    'bdquo':    '„',      # double low-9 quotation mark, U+201E NEW
    'beta':     'β',       # greek small letter beta, U+03B2 ISOgrk3
    'brvbar':   '\246',         # broken bar = broken vertical bar, U+00A6 ISOnum
    'bull':     '•',      # bullet = black small circle, U+2022 ISOpub
    'cap':      '∩',      # intersection = cap, U+2229 ISOtech
    'ccedil':   '\347',         # latin small letter c with cedilla, U+00E7 ISOlat1
    'cedil':    '\270',         # cedilla = spacing cedilla, U+00B8 ISOdia
    'cent':     '\242',         # cent sign, U+00A2 ISOnum
    'chi':      'χ',       # greek small letter chi, U+03C7 ISOgrk3
    'circ':     'ˆ',       # modifier letter circumflex accent, U+02C6 ISOpub
    'clubs':    '♣',      # black club suit = shamrock, U+2663 ISOpub
    'cong':     '≅',      # approximately equal to, U+2245 ISOtech
    'copy':     '\251',         # copyright sign, U+00A9 ISOnum
    'crarr':    '↵',      # downwards arrow with corner leftwards = carriage return, U+21B5 NEW
    'cup':      '∪',      # union = cup, U+222A ISOtech
    'curren':   '\244',         # currency sign, U+00A4 ISOnum
    'dArr':     '⇓',      # downwards double arrow, U+21D3 ISOamsa
    'dagger':   '†',      # dagger, U+2020 ISOpub
    'darr':     '↓',      # downwards arrow, U+2193 ISOnum
    'deg':      '\260',         # degree sign, U+00B0 ISOnum
    'delta':    'δ',       # greek small letter delta, U+03B4 ISOgrk3
    'diams':    '♦',      # black diamond suit, U+2666 ISOpub
    'divide':   '\367',         # division sign, U+00F7 ISOnum
    'eacute':   '\351',         # latin small letter e with acute, U+00E9 ISOlat1
    'ecirc':    '\352',         # latin small letter e with circumflex, U+00EA ISOlat1
    'egrave':   '\350',         # latin small letter e with grave, U+00E8 ISOlat1
    'empty':    '∅',      # empty set = null set = diameter, U+2205 ISOamso
    'emsp':     ' ',      # em space, U+2003 ISOpub
    'ensp':     ' ',      # en space, U+2002 ISOpub
    'epsilon':  'ε',       # greek small letter epsilon, U+03B5 ISOgrk3
    'equiv':    '≡',      # identical to, U+2261 ISOtech
    'eta':      'η',       # greek small letter eta, U+03B7 ISOgrk3
    'eth':      '\360',         # latin small letter eth, U+00F0 ISOlat1
    'euml':     '\353',         # latin small letter e with diaeresis, U+00EB ISOlat1
    'euro':     '€',      # euro sign, U+20AC NEW
    'exist':    '∃',      # there exists, U+2203 ISOtech
    'fnof':     'ƒ',       # latin small f with hook = function = florin, U+0192 ISOtech
    'forall':   '∀',      # for all, U+2200 ISOtech
    'frac12':   '\275',         # vulgar fraction one half = fraction one half, U+00BD ISOnum
    'frac14':   '\274',         # vulgar fraction one quarter = fraction one quarter, U+00BC ISOnum
    'frac34':   '\276',         # vulgar fraction three quarters = fraction three quarters, U+00BE ISOnum
    'frasl':    '⁄',      # fraction slash, U+2044 NEW
    'gamma':    'γ',       # greek small letter gamma, U+03B3 ISOgrk3
    'ge':       '≥',      # greater-than or equal to, U+2265 ISOtech
    'gt':       '\76',          # greater-than sign, U+003E ISOnum
    'hArr':     '⇔',      # left right double arrow, U+21D4 ISOamsa
    'harr':     '↔',      # left right arrow, U+2194 ISOamsa
    'hearts':   '♥',      # black heart suit = valentine, U+2665 ISOpub
    'hellip':   '…',      # horizontal ellipsis = three dot leader, U+2026 ISOpub
    'iacute':   '\355',         # latin small letter i with acute, U+00ED ISOlat1
    'icirc':    '\356',         # latin small letter i with circumflex, U+00EE ISOlat1
    'iexcl':    '\241',         # inverted exclamation mark, U+00A1 ISOnum
    'igrave':   '\354',         # latin small letter i with grave, U+00EC ISOlat1