2013-09-22 20:39:16 +00:00
|
|
|
# -*- coding: utf-8 -*-
|
|
|
|
"""
|
|
|
|
pygments.unistring
|
|
|
|
~~~~~~~~~~~~~~~~~~
|
|
|
|
|
|
|
|
Strings of all Unicode characters of a certain category.
|
|
|
|
Used for matching in Unicode-aware languages. Run to regenerate.
|
|
|
|
|
|
|
|
Inspired by chartypes_create.py from the MoinMoin project.
|
|
|
|
|
|
|
|
:copyright: Copyright 2006-2013 by the Pygments team, see AUTHORS.
|
|
|
|
:license: BSD, see LICENSE for details.
|
|
|
|
"""
|
|
|
|
from pygments.util import u_prefix
|
|
|
|
|
2013-09-22 21:48:55 +00:00
|
|
|
Cc = '\x00\x01\x02\x03\x04\x05\x06\x07\x08\t\n\x0b\x0c\r\x0e\x0f\x10\x11\x12\x13\x14\x15\x16\x17\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f\x7f\x80\x81\x82\x83\x84\x85\x86\x87\x88\x89\x8a\x8b\x8c\x8d\x8e\x8f\x90\x91\x92\x93\x94\x95\x96\x97\x98\x99\x9a\x9b\x9c\x9d\x9e\x9f'
|
2013-09-22 20:39:16 +00:00
|
|
|
|
2013-09-22 21:48:55 +00:00
|
|
|
Cf = '\xad\u0600\u0601\u0602\u0603\u06dd\u070f\u17b4\u17b5\u200b\u200c\u200d\u200e\u200f\u202a\u202b\u202c\u202d\u202e\u2060\u2061\u2062\u2063\u2064\u206a\u206b\u206c\u206d\u206e\u206f\ufeff\ufff9\ufffa\ufffb'
|
2013-09-22 20:39:16 +00:00
|
|
|
|
2013-09-22 21:48:55 +00:00
|
|
|
Cn = '\u0378\u0379\u037f\u0380\u0381\u0382\u0383\u038b\u038d\u03a2\u0526\u0527\u0528\u0529\u052a\u052b\u052c\u052d\u052e\u052f\u0530\u0557\u0558\u0560\u0588\u058b\u058c\u058d\u058e\u058f\u0590\u05c8\u05c9\u05ca\u05cb\u05cc\u05cd\u05ce\u05cf\u05eb\u05ec\u05ed\u05ee\u05ef\u05f5\u05f6\u05f7\u05f8\u05f9\u05fa\u05fb\u05fc\u05fd\u05fe\u05ff\u0604\u0605\u061c\u061d\u0620\u065f\u070e\u074b\u074c\u07b2\u07b3\u07b4\u07b5\u07b6\u07b7\u07b8\u07b9\u07ba\u07bb\u07bc\u07bd\u07be\u07bf\u07fb\u07fc\u07fd\u07fe\u07ff\u082e\u082f\u083f\u0840\u0841\u0842\u0843\u0844\u0845\u0846\u0847\u0848\u0849\u084a\u084b\u084c\u084d\u084e\u084f\u0850\u0851\u0852\u0853\u0854\u0855\u0856\u0857\u0858\u0859\u085a\u085b\u085c\u085d\u085e\u085f\u0860\u0861\u0862\u0863\u0864\u0865\u0866\u0867\u0868\u0869\u086a\u086b\u086c\u086d\u086e\u086f\u0870\u0871\u0872\u0873\u0874\u0875\u0876\u0877\u0878\u0879\u087a\u087b\u087c\u087d\u087e\u087f\u0880\u0881\u0882\u0883\u0884\u0885\u0886\u0887\u0888\u0889\u088a\u088b\u088c\u088d\u088e\u088f\u0890\u0891\u0892\u0893\u0894\u0895\u0896\u0897\u0898\u0899\u089a\u089b\u089c\u089d\u089e\u089f\u08a0\u08a1\u08a2\u08a3\u08a4\u08a5\u08a6\u08a7\u08a8\u08a9\u08aa\u08ab\u08ac\u08ad\u08ae\u08af\u08b0\u08b1\u08b2\u08b3\u08b4\u08b5\u08b6\u08b7\u08b8\u08b9\u08ba\u08bb\u08bc\u08bd\u08be\u08bf\u08c0\u08c1\u08c2\u08c3\u08c4\u08c5\u08c6\u08c7\u08c8\u08c9\u08ca\u08cb\u08cc\u08cd\u08ce\u08cf\u08d0\u08d1\u08d2\u08d3\u08d4\u08d5\u08d6\u08d7\u08d8\u08d9\u08da\u08db\u08dc\u08dd\u08de\u08df\u08e0\u08e1\u08e2\u08e3\u08e4\u08e5\u08e6\u08e7\u08e8\u08e9\u08ea\u08eb\u08ec\u08ed\u08ee\u08ef\u08f0\u08f1\u08f2\u08f3\u08f4\u08f5\u08f6\u08f7\u08f8\u08f9\u08fa\u08fb\u08fc\u08fd\u08fe\u08ff\u093a\u093b\u094f\u0956\u0957\u0973\u0974\u0975\u0976\u0977\u0978\u0980\u0984\u098d\u098e\u0991\u0992\u09a9\u09b1\u09b3\u09b4\u09b5\u09ba\u09bb\u09c5\u09c6\u09c9\u09ca\u09cf\u09d0\u09d1\u09d2\u09d3\u09d4\u09d5\u09d6\u09d8\u09d9\u09da\u09db\u09de\u09e4\u09e5\u09fc\u09fd\u09fe\u09ff\u0a00\u0a04\u0a0b\u0a0c\u0a0d\u0a0e\u0a11\u0a12\u0a29\u0a31\u0a34\u0a37\u0a3a\u0a3b\u0a3d\u0a43\u0a44\u0a45\u0a46\u0a49\u0a4a\u0a4e\u0a4f\u0a50\u0a52\u0a53\u0a54\u0a55\u0a56\u0a57\u0a58\u0a5d\u0a5f\u0a60\u0a61\u0a62\u0a63\u0a64\u0a65\u0a76\u0a77\u0a78\u0a79\u0a7a\u0a7b\u0a7c\u0a7d\u0a7e\u0a7f\u0a80\u0a84\u0a8e\u0a92\u0aa9\u0ab1\u0ab4\u0aba\u0abb\u0ac6\u0aca\u0ace\u0acf\u0ad1\u0ad2\u0ad3\u0ad4\u0ad5\u0ad6\u0ad7\u0ad8\u0ad9\u0ada\u0adb\u0adc\u0add\u0ade\u0adf\u0ae4\u0ae5\u0af0\u0af2\u0af3\u0af4\u0af5\u0af6\u0af7\u0af8\u0af9\u0afa\u0afb\u0afc\u0afd\u0afe\u0aff\u0b00\u0b04\u0b0d\u0b0e\u0b11\u0b12\u0b29\u0b31\u0b34\u0b3a\u0b3b\u0b45\u0b46\u0b49\u0b4a\u0b4e\u0b4f\u0b50\u0b51\u0b52\u0b53\u0b54\u0b55\u0b58\u0b59\u0b5a\u0b5b\u0b5e\u0b64\u0b65\u0b72\u0b73\u0b74\u0b75\u0b76\u0b77\u0b78\u0b79\u0b7a\u0b7b\u0b7c\u0b7d\u0b7e\u0b7f\u0b80\u0b81\u0b84\u0b8b\u0b8c\u0b8d\u0b91\u0b96\u0b97\u0b98\u0b9b\u0b9d\u0ba0\u0ba1\u0ba2\u0ba5\u0ba6\u0ba7\u0bab\u0bac\u0bad\u0bba\u0bbb\u0bbc\u0bbd\u0bc3\u0bc4\u0bc5\u0bc9\u0bce\u0bcf\u0bd1\u0bd2\u0bd3\u0bd4\u0bd5\u0bd6\u0bd8\u0bd9\u0bda\u0bdb\u0bdc\u0bdd\u0bde\u0bdf\u0be0\u0be1\u0be2\u0be3\u0be4\u0be5\u0bfb\u0bfc\u0bfd\u0bfe\u0bff\u0c00\u0c04\u0c0d\u0c11\u0c29\u0c34\u0c3a\u0c3b\u0c3c\u0c45\u0c49\u0c4e\u0c4f\u0c50\u0c51\u0c52\u0c53\u0c54\u0c57\u0c5a\u0c5b\u0c5c\u0c5d\u0c5e\u0c5f\u0c64\u0c65\u0c70\u0c71\u0c72\u0c73\u0c74\u0c75\u0c76\u0c77\u0c80\u0c81\u0c84\u0c8d\u0c91\u0ca9\u0cb4\u0cba\u0cbb\u0cc5\u0cc9\u0cce\u0ccf\u0cd0\u0cd1\u0cd2\u0cd3\u0cd4\u0cd7\u0cd8\u0cd9\u0cda\u0cdb\u0cdc\u0cdd\u0cdf\u0ce4\u0ce5\u0cf0\u0cf3\u0cf4\u0cf5\u0cf6\u0cf7\u0cf8\u0cf9\u0cfa\u0cfb\u0cfc\u0cfd\u0cfe\u0cff\u0d00\u0d01\u0d04\u0d0d\u0d11\u0d29\u0d3a\u0d3b\u0d3c\u0d45\u0d49\u0d4e\u0d4f\u0d50\u0d51\u0d52\u0d53\u0d54\u0d55\u0d56\u0d58\u0d59\u0d5a\u0d5b\u0d5c\u0d5d\u0d5e\u0d5f\u0d64\u0d65\u0d76\u0d77\u0d78\u0d80\u0d81\u0d84\u0d97\u0d98\u0d99\u0db2\u0dbc\u0dbe\u0dbf\u0dc7\u0dc8\u0dc9\u0dcb\u0dcc\u0dcd\u0dce\u0dd5\u0dd7\u0de0\u0de1\u0de2\u0de3\u0de4\u0de5\u0de6\u0de7\u0de8\u0de9\u0dea\u0deb\u0dec\u0ded\u0dee\u0def\u0df0\u0df1\u0df5\u0df6\u0df7\u0df8\u0df9\u0dfa\u0dfb\u0dfc\u0dfd\u0dfe\u0dff\u0e00\u0e3b\u0e3c\u0e3d\u0
|
2013-09-22 20:39:16 +00:00
|
|
|
|
2013-09-22 21:48:55 +00:00
|
|
|
Co = '\ue000\ue001\ue002\ue003\ue004\ue005\ue006\ue007\ue008\ue009\ue00a\ue00b\ue00c\ue00d\ue00e\ue00f\ue010\ue011\ue012\ue013\ue014\ue015\ue016\ue017\ue018\ue019\ue01a\ue01b\ue01c\ue01d\ue01e\ue01f\ue020\ue021\ue022\ue023\ue024\ue025\ue026\ue027\ue028\ue029\ue02a\ue02b\ue02c\ue02d\ue02e\ue02f\ue030\ue031\ue032\ue033\ue034\ue035\ue036\ue037\ue038\ue039\ue03a\ue03b\ue03c\ue03d\ue03e\ue03f\ue040\ue041\ue042\ue043\ue044\ue045\ue046\ue047\ue048\ue049\ue04a\ue04b\ue04c\ue04d\ue04e\ue04f\ue050\ue051\ue052\ue053\ue054\ue055\ue056\ue057\ue058\ue059\ue05a\ue05b\ue05c\ue05d\ue05e\ue05f\ue060\ue061\ue062\ue063\ue064\ue065\ue066\ue067\ue068\ue069\ue06a\ue06b\ue06c\ue06d\ue06e\ue06f\ue070\ue071\ue072\ue073\ue074\ue075\ue076\ue077\ue078\ue079\ue07a\ue07b\ue07c\ue07d\ue07e\ue07f\ue080\ue081\ue082\ue083\ue084\ue085\ue086\ue087\ue088\ue089\ue08a\ue08b\ue08c\ue08d\ue08e\ue08f\ue090\ue091\ue092\ue093\ue094\ue095\ue096\ue097\ue098\ue099\ue09a\ue09b\ue09c\ue09d\ue09e\ue09f\ue0a0\ue0a1\ue0a2\ue0a3\ue0a4\ue0a5\ue0a6\ue0a7\ue0a8\ue0a9\ue0aa\ue0ab\ue0ac\ue0ad\ue0ae\ue0af\ue0b0\ue0b1\ue0b2\ue0b3\ue0b4\ue0b5\ue0b6\ue0b7\ue0b8\ue0b9\ue0ba\ue0bb\ue0bc\ue0bd\ue0be\ue0bf\ue0c0\ue0c1\ue0c2\ue0c3\ue0c4\ue0c5\ue0c6\ue0c7\ue0c8\ue0c9\ue0ca\ue0cb\ue0cc\ue0cd\ue0ce\ue0cf\ue0d0\ue0d1\ue0d2\ue0d3\ue0d4\ue0d5\ue0d6\ue0d7\ue0d8\ue0d9\ue0da\ue0db\ue0dc\ue0dd\ue0de\ue0df\ue0e0\ue0e1\ue0e2\ue0e3\ue0e4\ue0e5\ue0e6\ue0e7\ue0e8\ue0e9\ue0ea\ue0eb\ue0ec\ue0ed\ue0ee\ue0ef\ue0f0\ue0f1\ue0f2\ue0f3\ue0f4\ue0f5\ue0f6\ue0f7\ue0f8\ue0f9\ue0fa\ue0fb\ue0fc\ue0fd\ue0fe\ue0ff\ue100\ue101\ue102\ue103\ue104\ue105\ue106\ue107\ue108\ue109\ue10a\ue10b\ue10c\ue10d\ue10e\ue10f\ue110\ue111\ue112\ue113\ue114\ue115\ue116\ue117\ue118\ue119\ue11a\ue11b\ue11c\ue11d\ue11e\ue11f\ue120\ue121\ue122\ue123\ue124\ue125\ue126\ue127\ue128\ue129\ue12a\ue12b\ue12c\ue12d\ue12e\ue12f\ue130\ue131\ue132\ue133\ue134\ue135\ue136\ue137\ue138\ue139\ue13a\ue13b\ue13c\ue13d\ue13e\ue13f\ue140\ue141\ue142\ue143\ue144\ue145\ue146\ue147\ue148\ue149\ue14a\ue14b\ue14c\ue14d\ue14e\ue14f\ue150\ue151\ue152\ue153\ue154\ue155\ue156\ue157\ue158\ue159\ue15a\ue15b\ue15c\ue15d\ue15e\ue15f\ue160\ue161\ue162\ue163\ue164\ue165\ue166\ue167\ue168\ue169\ue16a\ue16b\ue16c\ue16d\ue16e\ue16f\ue170\ue171\ue172\ue173\ue174\ue175\ue176\ue177\ue178\ue179\ue17a\ue17b\ue17c\ue17d\ue17e\ue17f\ue180\ue181\ue182\ue183\ue184\ue185\ue186\ue187\ue188\ue189\ue18a\ue18b\ue18c\ue18d\ue18e\ue18f\ue190\ue191\ue192\ue193\ue194\ue195\ue196\ue197\ue198\ue199\ue19a\ue19b\ue19c\ue19d\ue19e\ue19f\ue1a0\ue1a1\ue1a2\ue1a3\ue1a4\ue1a5\ue1a6\ue1a7\ue1a8\ue1a9\ue1aa\ue1ab\ue1ac\ue1ad\ue1ae\ue1af\ue1b0\ue1b1\ue1b2\ue1b3\ue1b4\ue1b5\ue1b6\ue1b7\ue1b8\ue1b9\ue1ba\ue1bb\ue1bc\ue1bd\ue1be\ue1bf\ue1c0\ue1c1\ue1c2\ue1c3\ue1c4\ue1c5\ue1c6\ue1c7\ue1c8\ue1c9\ue1ca\ue1cb\ue1cc\ue1cd\ue1ce\ue1cf\ue1d0\ue1d1\ue1d2\ue1d3\ue1d4\ue1d5\ue1d6\ue1d7\ue1d8\ue1d9\ue1da\ue1db\ue1dc\ue1dd\ue1de\ue1df\ue1e0\ue1e1\ue1e2\ue1e3\ue1e4\ue1e5\ue1e6\ue1e7\ue1e8\ue1e9\ue1ea\ue1eb\ue1ec\ue1ed\ue1ee\ue1ef\ue1f0\ue1f1\ue1f2\ue1f3\ue1f4\ue1f5\ue1f6\ue1f7\ue1f8\ue1f9\ue1fa\ue1fb\ue1fc\ue1fd\ue1fe\ue1ff\ue200\ue201\ue202\ue203\ue204\ue205\ue206\ue207\ue208\ue209\ue20a\ue20b\ue20c\ue20d\ue20e\ue20f\ue210\ue211\ue212\ue213\ue214\ue215\ue216\ue217\ue218\ue219\ue21a\ue21b\ue21c\ue21d\ue21e\ue21f\ue220\ue221\ue222\ue223\ue224\ue225\ue226\ue227\ue228\ue229\ue22a\ue22b\ue22c\ue22d\ue22e\ue22f\ue230\ue231\ue232\ue233\ue234\ue235\ue236\ue237\ue238\ue239\ue23a\ue23b\ue23c\ue23d\ue23e\ue23f\ue240\ue241\ue242\ue243\ue244\ue245\ue246\ue247\ue248\ue249\ue24a\ue24b\ue24c\ue24d\ue24e\ue24f\ue250\ue251\ue252\ue253\ue254\ue255\ue256\ue257\ue258\ue259\ue25a\ue25b\ue25c\ue25d\ue25e\ue25f\ue260\ue261\ue262\ue263\ue264\ue265\ue266\ue267\ue268\ue269\ue26a\ue26b\ue26c\ue26d\ue26e\ue26f\ue270\ue271\ue272\ue273\ue274\ue275\ue276\ue277\ue278\ue279\ue27a\ue27b\ue27c\ue27d\ue27e\ue27f\ue280\ue281\ue282\ue283\ue284\ue285\ue286\ue287\ue288\ue289\ue28a\ue28b\ue28c\ue28d\ue28e\ue28f\ue290\ue291\ue292\ue293\ue294\ue295\ue296\ue297\ue298\ue299\ue29a\ue29b\ue29c\ue29d\ue29e\ue29f\ue2a0\ue2a1\ue2a2\ue2a3\ue2a4\ue2a5\ue2a6\ue2a7\ue2a8\ue
|
2013-09-22 20:39:16 +00:00
|
|
|
|
|
|
|
try:
|
|
|
|
Cs = eval(u_prefix + r"'\ud800\ud801\ud802\ud803\ud804\ud805\ud806\ud807\ud808\ud809\ud80a\ud80b\ud80c\ud80d\ud80e\ud80f\ud810\ud811\ud812\ud813\ud814\ud815\ud816\ud817\ud818\ud819\ud81a\ud81b\ud81c\ud81d\ud81e\ud81f\ud820\ud821\ud822\ud823\ud824\ud825\ud826\ud827\ud828\ud829\ud82a\ud82b\ud82c\ud82d\ud82e\ud82f\ud830\ud831\ud832\ud833\ud834\ud835\ud836\ud837\ud838\ud839\ud83a\ud83b\ud83c\ud83d\ud83e\ud83f\ud840\ud841\ud842\ud843\ud844\ud845\ud846\ud847\ud848\ud849\ud84a\ud84b\ud84c\ud84d\ud84e\ud84f\ud850\ud851\ud852\ud853\ud854\ud855\ud856\ud857\ud858\ud859\ud85a\ud85b\ud85c\ud85d\ud85e\ud85f\ud860\ud861\ud862\ud863\ud864\ud865\ud866\ud867\ud868\ud869\ud86a\ud86b\ud86c\ud86d\ud86e\ud86f\ud870\ud871\ud872\ud873\ud874\ud875\ud876\ud877\ud878\ud879\ud87a\ud87b\ud87c\ud87d\ud87e\ud87f\ud880\ud881\ud882\ud883\ud884\ud885\ud886\ud887\ud888\ud889\ud88a\ud88b\ud88c\ud88d\ud88e\ud88f\ud890\ud891\ud892\ud893\ud894\ud895\ud896\ud897\ud898\ud899\ud89a\ud89b\ud89c\ud89d\ud89e\ud89f\ud8a0\ud8a1\ud8a2\ud8a3\ud8a4\ud8a5\ud8a6\ud8a7\ud8a8\ud8a9\ud8aa\ud8ab\ud8ac\ud8ad\ud8ae\ud8af\ud8b0\ud8b1\ud8b2\ud8b3\ud8b4\ud8b5\ud8b6\ud8b7\ud8b8\ud8b9\ud8ba\ud8bb\ud8bc\ud8bd\ud8be\ud8bf\ud8c0\ud8c1\ud8c2\ud8c3\ud8c4\ud8c5\ud8c6\ud8c7\ud8c8\ud8c9\ud8ca\ud8cb\ud8cc\ud8cd\ud8ce\ud8cf\ud8d0\ud8d1\ud8d2\ud8d3\ud8d4\ud8d5\ud8d6\ud8d7\ud8d8\ud8d9\ud8da\ud8db\ud8dc\ud8dd\ud8de\ud8df\ud8e0\ud8e1\ud8e2\ud8e3\ud8e4\ud8e5\ud8e6\ud8e7\ud8e8\ud8e9\ud8ea\ud8eb\ud8ec\ud8ed\ud8ee\ud8ef\ud8f0\ud8f1\ud8f2\ud8f3\ud8f4\ud8f5\ud8f6\ud8f7\ud8f8\ud8f9\ud8fa\ud8fb\ud8fc\ud8fd\ud8fe\ud8ff\ud900\ud901\ud902\ud903\ud904\ud905\ud906\ud907\ud908\ud909\ud90a\ud90b\ud90c\ud90d\ud90e\ud90f\ud910\ud911\ud912\ud913\ud914\ud915\ud916\ud917\ud918\ud919\ud91a\ud91b\ud91c\ud91d\ud91e\ud91f\ud920\ud921\ud922\ud923\ud924\ud925\ud926\ud927\ud928\ud929\ud92a\ud92b\ud92c\ud92d\ud92e\ud92f\ud930\ud931\ud932\ud933\ud934\ud935\ud936\ud937\ud938\ud939\ud93a\ud93b\ud93c\ud93d\ud93e\ud93f\ud940\ud941\ud942\ud943\ud944\ud945\ud946\ud947\ud948\ud949\ud94a\ud94b\ud94c\ud94d\ud94e\ud94f\ud950\ud951\ud952\ud953\ud954\ud955\ud956\ud957\ud958\ud959\ud95a\ud95b\ud95c\ud95d\ud95e\ud95f\ud960\ud961\ud962\ud963\ud964\ud965\ud966\ud967\ud968\ud969\ud96a\ud96b\ud96c\ud96d\ud96e\ud96f\ud970\ud971\ud972\ud973\ud974\ud975\ud976\ud977\ud978\ud979\ud97a\ud97b\ud97c\ud97d\ud97e\ud97f\ud980\ud981\ud982\ud983\ud984\ud985\ud986\ud987\ud988\ud989\ud98a\ud98b\ud98c\ud98d\ud98e\ud98f\ud990\ud991\ud992\ud993\ud994\ud995\ud996\ud997\ud998\ud999\ud99a\ud99b\ud99c\ud99d\ud99e\ud99f\ud9a0\ud9a1\ud9a2\ud9a3\ud9a4\ud9a5\ud9a6\ud9a7\ud9a8\ud9a9\ud9aa\ud9ab\ud9ac\ud9ad\ud9ae\ud9af\ud9b0\ud9b1\ud9b2\ud9b3\ud9b4\ud9b5\ud9b6\ud9b7\ud9b8\ud9b9\ud9ba\ud9bb\ud9bc\ud9bd\ud9be\ud9bf\ud9c0\ud9c1\ud9c2\ud9c3\ud9c4\ud9c5\ud9c6\ud9c7\ud9c8\ud9c9\ud9ca\ud9cb\ud9cc\ud9cd\ud9ce\ud9cf\ud9d0\ud9d1\ud9d2\ud9d3\ud9d4\ud9d5\ud9d6\ud9d7\ud9d8\ud9d9\ud9da\ud9db\ud9dc\ud9dd\ud9de\ud9df\ud9e0\ud9e1\ud9e2\ud9e3\ud9e4\ud9e5\ud9e6\ud9e7\ud9e8\ud9e9\ud9ea\ud9eb\ud9ec\ud9ed\ud9ee\ud9ef\ud9f0\ud9f1\ud9f2\ud9f3\ud9f4\ud9f5\ud9f6\ud9f7\ud9f8\ud9f9\ud9fa\ud9fb\ud9fc\ud9fd\ud9fe\ud9ff\uda00\uda01\uda02\uda03\uda04\uda05\uda06\uda07\uda08\uda09\uda0a\uda0b\uda0c\uda0d\uda0e\uda0f\uda10\uda11\uda12\uda13\uda14\uda15\uda16\uda17\uda18\uda19\uda1a\uda1b\uda1c\uda1d\uda1e\uda1f\uda20\uda21\uda22\uda23\uda24\uda25\uda26\uda27\uda28\uda29\uda2a\uda2b\uda2c\uda2d\uda2e\uda2f\uda30\uda31\uda32\uda33\uda34\uda35\uda36\uda37\uda38\uda39\uda3a\uda3b\uda3c\uda3d\uda3e\uda3f\uda40\uda41\uda42\uda43\uda44\uda45\uda46\uda47\uda48\uda49\uda4a\uda4b\uda4c\uda4d\uda4e\uda4f\uda50\uda51\uda52\uda53\uda54\uda55\uda56\uda57\uda58\uda59\uda5a\uda5b\uda5c\uda5d\uda5e\uda5f\uda60\uda61\uda62\uda63\uda64\uda65\uda66\uda67\uda68\uda69\uda6a\uda6b\uda6c\uda6d\uda6e\uda6f\uda70\uda71\uda72\uda73\uda74\uda75\uda76\uda77\uda78\uda79\uda7a\uda7b\uda7c\uda7d\uda7e\uda7f\uda80\uda81\uda82\uda83\uda84\uda85\uda86\uda87\uda88\uda89\uda8a\uda8b\uda8c\uda8d\uda8e\uda8f\uda90\uda91\uda92\uda93\uda94\uda95\uda96\uda97\uda98\uda99\uda9a\uda9b\uda9c\uda9d\uda9e\uda9f\udaa0\udaa1\udaa2\udaa3\udaa4\udaa
|
|
|
|
except UnicodeDecodeError:
|
|
|
|
Cs = '' # Jython can't handle isolated surrogates
|
|
|
|
|
2013-09-22 21:48:55 +00:00
|
|
|
Ll = 'abcdefghijklmnopqrstuvwxyz\xaa\xb5\xba\xdf\xe0\xe1\xe2\xe3\xe4\xe5\xe6\xe7\xe8\xe9\xea\xeb\xec\xed\xee\xef\xf0\xf1\xf2\xf3\xf4\xf5\xf6\xf8\xf9\xfa\xfb\xfc\xfd\xfe\xff\u0101\u0103\u0105\u0107\u0109\u010b\u010d\u010f\u0111\u0113\u0115\u0117\u0119\u011b\u011d\u011f\u0121\u0123\u0125\u0127\u0129\u012b\u012d\u012f\u0131\u0133\u0135\u0137\u0138\u013a\u013c\u013e\u0140\u0142\u0144\u0146\u0148\u0149\u014b\u014d\u014f\u0151\u0153\u0155\u0157\u0159\u015b\u015d\u015f\u0161\u0163\u0165\u0167\u0169\u016b\u016d\u016f\u0171\u0173\u0175\u0177\u017a\u017c\u017e\u017f\u0180\u0183\u0185\u0188\u018c\u018d\u0192\u0195\u0199\u019a\u019b\u019e\u01a1\u01a3\u01a5\u01a8\u01aa\u01ab\u01ad\u01b0\u01b4\u01b6\u01b9\u01ba\u01bd\u01be\u01bf\u01c6\u01c9\u01cc\u01ce\u01d0\u01d2\u01d4\u01d6\u01d8\u01da\u01dc\u01dd\u01df\u01e1\u01e3\u01e5\u01e7\u01e9\u01eb\u01ed\u01ef\u01f0\u01f3\u01f5\u01f9\u01fb\u01fd\u01ff\u0201\u0203\u0205\u0207\u0209\u020b\u020d\u020f\u0211\u0213\u0215\u0217\u0219\u021b\u021d\u021f\u0221\u0223\u0225\u0227\u0229\u022b\u022d\u022f\u0231\u0233\u0234\u0235\u0236\u0237\u0238\u0239\u023c\u023f\u0240\u0242\u0247\u0249\u024b\u024d\u024f\u0250\u0251\u0252\u0253\u0254\u0255\u0256\u0257\u0258\u0259\u025a\u025b\u025c\u025d\u025e\u025f\u0260\u0261\u0262\u0263\u0264\u0265\u0266\u0267\u0268\u0269\u026a\u026b\u026c\u026d\u026e\u026f\u0270\u0271\u0272\u0273\u0274\u0275\u0276\u0277\u0278\u0279\u027a\u027b\u027c\u027d\u027e\u027f\u0280\u0281\u0282\u0283\u0284\u0285\u0286\u0287\u0288\u0289\u028a\u028b\u028c\u028d\u028e\u028f\u0290\u0291\u0292\u0293\u0295\u0296\u0297\u0298\u0299\u029a\u029b\u029c\u029d\u029e\u029f\u02a0\u02a1\u02a2\u02a3\u02a4\u02a5\u02a6\u02a7\u02a8\u02a9\u02aa\u02ab\u02ac\u02ad\u02ae\u02af\u0371\u0373\u0377\u037b\u037c\u037d\u0390\u03ac\u03ad\u03ae\u03af\u03b0\u03b1\u03b2\u03b3\u03b4\u03b5\u03b6\u03b7\u03b8\u03b9\u03ba\u03bb\u03bc\u03bd\u03be\u03bf\u03c0\u03c1\u03c2\u03c3\u03c4\u03c5\u03c6\u03c7\u03c8\u03c9\u03ca\u03cb\u03cc\u03cd\u03ce\u03d0\u03d1\u03d5\u03d6\u03d7\u03d9\u03db\u03dd\u03df\u03e1\u03e3\u03e5\u03e7\u03e9\u03eb\u03ed\u03ef\u03f0\u03f1\u03f2\u03f3\u03f5\u03f8\u03fb\u03fc\u0430\u0431\u0432\u0433\u0434\u0435\u0436\u0437\u0438\u0439\u043a\u043b\u043c\u043d\u043e\u043f\u0440\u0441\u0442\u0443\u0444\u0445\u0446\u0447\u0448\u0449\u044a\u044b\u044c\u044d\u044e\u044f\u0450\u0451\u0452\u0453\u0454\u0455\u0456\u0457\u0458\u0459\u045a\u045b\u045c\u045d\u045e\u045f\u0461\u0463\u0465\u0467\u0469\u046b\u046d\u046f\u0471\u0473\u0475\u0477\u0479\u047b\u047d\u047f\u0481\u048b\u048d\u048f\u0491\u0493\u0495\u0497\u0499\u049b\u049d\u049f\u04a1\u04a3\u04a5\u04a7\u04a9\u04ab\u04ad\u04af\u04b1\u04b3\u04b5\u04b7\u04b9\u04bb\u04bd\u04bf\u04c2\u04c4\u04c6\u04c8\u04ca\u04cc\u04ce\u04cf\u04d1\u04d3\u04d5\u04d7\u04d9\u04db\u04dd\u04df\u04e1\u04e3\u04e5\u04e7\u04e9\u04eb\u04ed\u04ef\u04f1\u04f3\u04f5\u04f7\u04f9\u04fb\u04fd\u04ff\u0501\u0503\u0505\u0507\u0509\u050b\u050d\u050f\u0511\u0513\u0515\u0517\u0519\u051b\u051d\u051f\u0521\u0523\u0525\u0561\u0562\u0563\u0564\u0565\u0566\u0567\u0568\u0569\u056a\u056b\u056c\u056d\u056e\u056f\u0570\u0571\u0572\u0573\u0574\u0575\u0576\u0577\u0578\u0579\u057a\u057b\u057c\u057d\u057e\u057f\u0580\u0581\u0582\u0583\u0584\u0585\u0586\u0587\u1d00\u1d01\u1d02\u1d03\u1d04\u1d05\u1d06\u1d07\u1d08\u1d09\u1d0a\u1d0b\u1d0c\u1d0d\u1d0e\u1d0f\u1d10\u1d11\u1d12\u1d13\u1d14\u1d15\u1d16\u1d17\u1d18\u1d19\u1d1a\u1d1b\u1d1c\u1d1d\u1d1e\u1d1f\u1d20\u1d21\u1d22\u1d23\u1d24\u1d25\u1d26\u1d27\u1d28\u1d29\u1d2a\u1d2b\u1d62\u1d63\u1d64\u1d65\u1d66\u1d67\u1d68\u1d69\u1d6a\u1d6b\u1d6c\u1d6d\u1d6e\u1d6f\u1d70\u1d71\u1d72\u1d73\u1d74\u1d75\u1d76\u1d77\u1d79\u1d7a\u1d7b\u1d7c\u1d7d\u1d7e\u1d7f\u1d80\u1d81\u1d82\u1d83\u1d84\u1d85\u1d86\u1d87\u1d88\u1d89\u1d8a\u1d8b\u1d8c\u1d8d\u1d8e\u1d8f\u1d90\u1d91\u1d92\u1d93\u1d94\u1d95\u1d96\u1d97\u1d98\u1d99\u1d9a\u1e01\u1e03\u1e05\u1e07\u1e09\u1e0b\u1e0d\u1e0f\u1e11\u1e13\u1e15\u1e17\u1e19\u1e1b\u1e1d\u1e1f\u1e21\u1e23\u1e25\u1e27\u1e29\u1e2b\u1e2d\u1e2f\u1e31\u1e33\u1e35\u1e37\u1e39\u1e3b\u1e3d\u1e3f\u1e41\u1e43\u1e45\u1e47\u1e49\u1e4b\u1e4d\u1e4f\u1e51\u1e53\u1e55\u1e57\u1e59\u1e5b\u1e5d\u1e5f\u1e6
|
2013-09-22 20:39:16 +00:00
|
|
|
|
2013-09-22 21:48:55 +00:00
|
|
|
Lm = '\u02b0\u02b1\u02b2\u02b3\u02b4\u02b5\u02b6\u02b7\u02b8\u02b9\u02ba\u02bb\u02bc\u02bd\u02be\u02bf\u02c0\u02c1\u02c6\u02c7\u02c8\u02c9\u02ca\u02cb\u02cc\u02cd\u02ce\u02cf\u02d0\u02d1\u02e0\u02e1\u02e2\u02e3\u02e4\u02ec\u02ee\u0374\u037a\u0559\u0640\u06e5\u06e6\u07f4\u07f5\u07fa\u081a\u0824\u0828\u0971\u0e46\u0ec6\u10fc\u17d7\u1843\u1aa7\u1c78\u1c79\u1c7a\u1c7b\u1c7c\u1c7d\u1d2c\u1d2d\u1d2e\u1d2f\u1d30\u1d31\u1d32\u1d33\u1d34\u1d35\u1d36\u1d37\u1d38\u1d39\u1d3a\u1d3b\u1d3c\u1d3d\u1d3e\u1d3f\u1d40\u1d41\u1d42\u1d43\u1d44\u1d45\u1d46\u1d47\u1d48\u1d49\u1d4a\u1d4b\u1d4c\u1d4d\u1d4e\u1d4f\u1d50\u1d51\u1d52\u1d53\u1d54\u1d55\u1d56\u1d57\u1d58\u1d59\u1d5a\u1d5b\u1d5c\u1d5d\u1d5e\u1d5f\u1d60\u1d61\u1d78\u1d9b\u1d9c\u1d9d\u1d9e\u1d9f\u1da0\u1da1\u1da2\u1da3\u1da4\u1da5\u1da6\u1da7\u1da8\u1da9\u1daa\u1dab\u1dac\u1dad\u1dae\u1daf\u1db0\u1db1\u1db2\u1db3\u1db4\u1db5\u1db6\u1db7\u1db8\u1db9\u1dba\u1dbb\u1dbc\u1dbd\u1dbe\u1dbf\u2071\u207f\u2090\u2091\u2092\u2093\u2094\u2c7d\u2d6f\u2e2f\u3005\u3031\u3032\u3033\u3034\u3035\u303b\u309d\u309e\u30fc\u30fd\u30fe\ua015\ua4f8\ua4f9\ua4fa\ua4fb\ua4fc\ua4fd\ua60c\ua67f\ua717\ua718\ua719\ua71a\ua71b\ua71c\ua71d\ua71e\ua71f\ua770\ua788\ua9cf\uaa70\uaadd\uff70\uff9e\uff9f'
|
2013-09-22 20:39:16 +00:00
|
|
|
|
2013-09-22 21:48:55 +00:00
|
|
|
Lo = '\u01bb\u01c0\u01c1\u01c2\u01c3\u0294\u05d0\u05d1\u05d2\u05d3\u05d4\u05d5\u05d6\u05d7\u05d8\u05d9\u05da\u05db\u05dc\u05dd\u05de\u05df\u05e0\u05e1\u05e2\u05e3\u05e4\u05e5\u05e6\u05e7\u05e8\u05e9\u05ea\u05f0\u05f1\u05f2\u0621\u0622\u0623\u0624\u0625\u0626\u0627\u0628\u0629\u062a\u062b\u062c\u062d\u062e\u062f\u0630\u0631\u0632\u0633\u0634\u0635\u0636\u0637\u0638\u0639\u063a\u063b\u063c\u063d\u063e\u063f\u0641\u0642\u0643\u0644\u0645\u0646\u0647\u0648\u0649\u064a\u066e\u066f\u0671\u0672\u0673\u0674\u0675\u0676\u0677\u0678\u0679\u067a\u067b\u067c\u067d\u067e\u067f\u0680\u0681\u0682\u0683\u0684\u0685\u0686\u0687\u0688\u0689\u068a\u068b\u068c\u068d\u068e\u068f\u0690\u0691\u0692\u0693\u0694\u0695\u0696\u0697\u0698\u0699\u069a\u069b\u069c\u069d\u069e\u069f\u06a0\u06a1\u06a2\u06a3\u06a4\u06a5\u06a6\u06a7\u06a8\u06a9\u06aa\u06ab\u06ac\u06ad\u06ae\u06af\u06b0\u06b1\u06b2\u06b3\u06b4\u06b5\u06b6\u06b7\u06b8\u06b9\u06ba\u06bb\u06bc\u06bd\u06be\u06bf\u06c0\u06c1\u06c2\u06c3\u06c4\u06c5\u06c6\u06c7\u06c8\u06c9\u06ca\u06cb\u06cc\u06cd\u06ce\u06cf\u06d0\u06d1\u06d2\u06d3\u06d5\u06ee\u06ef\u06fa\u06fb\u06fc\u06ff\u0710\u0712\u0713\u0714\u0715\u0716\u0717\u0718\u0719\u071a\u071b\u071c\u071d\u071e\u071f\u0720\u0721\u0722\u0723\u0724\u0725\u0726\u0727\u0728\u0729\u072a\u072b\u072c\u072d\u072e\u072f\u074d\u074e\u074f\u0750\u0751\u0752\u0753\u0754\u0755\u0756\u0757\u0758\u0759\u075a\u075b\u075c\u075d\u075e\u075f\u0760\u0761\u0762\u0763\u0764\u0765\u0766\u0767\u0768\u0769\u076a\u076b\u076c\u076d\u076e\u076f\u0770\u0771\u0772\u0773\u0774\u0775\u0776\u0777\u0778\u0779\u077a\u077b\u077c\u077d\u077e\u077f\u0780\u0781\u0782\u0783\u0784\u0785\u0786\u0787\u0788\u0789\u078a\u078b\u078c\u078d\u078e\u078f\u0790\u0791\u0792\u0793\u0794\u0795\u0796\u0797\u0798\u0799\u079a\u079b\u079c\u079d\u079e\u079f\u07a0\u07a1\u07a2\u07a3\u07a4\u07a5\u07b1\u07ca\u07cb\u07cc\u07cd\u07ce\u07cf\u07d0\u07d1\u07d2\u07d3\u07d4\u07d5\u07d6\u07d7\u07d8\u07d9\u07da\u07db\u07dc\u07dd\u07de\u07df\u07e0\u07e1\u07e2\u07e3\u07e4\u07e5\u07e6\u07e7\u07e8\u07e9\u07ea\u0800\u0801\u0802\u0803\u0804\u0805\u0806\u0807\u0808\u0809\u080a\u080b\u080c\u080d\u080e\u080f\u0810\u0811\u0812\u0813\u0814\u0815\u0904\u0905\u0906\u0907\u0908\u0909\u090a\u090b\u090c\u090d\u090e\u090f\u0910\u0911\u0912\u0913\u0914\u0915\u0916\u0917\u0918\u0919\u091a\u091b\u091c\u091d\u091e\u091f\u0920\u0921\u0922\u0923\u0924\u0925\u0926\u0927\u0928\u0929\u092a\u092b\u092c\u092d\u092e\u092f\u0930\u0931\u0932\u0933\u0934\u0935\u0936\u0937\u0938\u0939\u093d\u0950\u0958\u0959\u095a\u095b\u095c\u095d\u095e\u095f\u0960\u0961\u0972\u0979\u097a\u097b\u097c\u097d\u097e\u097f\u0985\u0986\u0987\u0988\u0989\u098a\u098b\u098c\u098f\u0990\u0993\u0994\u0995\u0996\u0997\u0998\u0999\u099a\u099b\u099c\u099d\u099e\u099f\u09a0\u09a1\u09a2\u09a3\u09a4\u09a5\u09a6\u09a7\u09a8\u09aa\u09ab\u09ac\u09ad\u09ae\u09af\u09b0\u09b2\u09b6\u09b7\u09b8\u09b9\u09bd\u09ce\u09dc\u09dd\u09df\u09e0\u09e1\u09f0\u09f1\u0a05\u0a06\u0a07\u0a08\u0a09\u0a0a\u0a0f\u0a10\u0a13\u0a14\u0a15\u0a16\u0a17\u0a18\u0a19\u0a1a\u0a1b\u0a1c\u0a1d\u0a1e\u0a1f\u0a20\u0a21\u0a22\u0a23\u0a24\u0a25\u0a26\u0a27\u0a28\u0a2a\u0a2b\u0a2c\u0a2d\u0a2e\u0a2f\u0a30\u0a32\u0a33\u0a35\u0a36\u0a38\u0a39\u0a59\u0a5a\u0a5b\u0a5c\u0a5e\u0a72\u0a73\u0a74\u0a85\u0a86\u0a87\u0a88\u0a89\u0a8a\u0a8b\u0a8c\u0a8d\u0a8f\u0a90\u0a91\u0a93\u0a94\u0a95\u0a96\u0a97\u0a98\u0a99\u0a9a\u0a9b\u0a9c\u0a9d\u0a9e\u0a9f\u0aa0\u0aa1\u0aa2\u0aa3\u0aa4\u0aa5\u0aa6\u0aa7\u0aa8\u0aaa\u0aab\u0aac\u0aad\u0aae\u0aaf\u0ab0\u0ab2\u0ab3\u0ab5\u0ab6\u0ab7\u0ab8\u0ab9\u0abd\u0ad0\u0ae0\u0ae1\u0b05\u0b06\u0b07\u0b08\u0b09\u0b0a\u0b0b\u0b0c\u0b0f\u0b10\u0b13\u0b14\u0b15\u0b16\u0b17\u0b18\u0b19\u0b1a\u0b1b\u0b1c\u0b1d\u0b1e\u0b1f\u0b20\u0b21\u0b22\u0b23\u0b24\u0b25\u0b26\u0b27\u0b28\u0b2a\u0b2b\u0b2c\u0b2d\u0b2e\u0b2f\u0b30\u0b32\u0b33\u0b35\u0b36\u0b37\u0b38\u0b39\u0b3d\u0b5c\u0b5d\u0b5f\u0b60\u0b61\u0b71\u0b83\u0b85\u0b86\u0b87\u0b88\u0b89\u0b8a\u0b8e\u0b8f\u0b90\u0b92\u0b93\u0b94\u0b95\u0b99\u0b9a\u0b9c\u0b9e\u0b9f\u0ba3\u0ba4\u0ba8\u0ba9\u0baa\u0bae\u0baf\u0bb0\u0bb1\u0bb2\u0bb3\u0bb4\u0bb5\u0bb6\u0bb7\u0bb8\u0bb9\u0bd0\u0
|
2013-09-22 20:39:16 +00:00
|
|
|
|
2013-09-22 21:48:55 +00:00
|
|
|
Lt = '\u01c5\u01c8\u01cb\u01f2\u1f88\u1f89\u1f8a\u1f8b\u1f8c\u1f8d\u1f8e\u1f8f\u1f98\u1f99\u1f9a\u1f9b\u1f9c\u1f9d\u1f9e\u1f9f\u1fa8\u1fa9\u1faa\u1fab\u1fac\u1fad\u1fae\u1faf\u1fbc\u1fcc\u1ffc'
|
2013-09-22 20:39:16 +00:00
|
|
|
|
2013-09-22 21:48:55 +00:00
|
|
|
Lu = 'ABCDEFGHIJKLMNOPQRSTUVWXYZ\xc0\xc1\xc2\xc3\xc4\xc5\xc6\xc7\xc8\xc9\xca\xcb\xcc\xcd\xce\xcf\xd0\xd1\xd2\xd3\xd4\xd5\xd6\xd8\xd9\xda\xdb\xdc\xdd\xde\u0100\u0102\u0104\u0106\u0108\u010a\u010c\u010e\u0110\u0112\u0114\u0116\u0118\u011a\u011c\u011e\u0120\u0122\u0124\u0126\u0128\u012a\u012c\u012e\u0130\u0132\u0134\u0136\u0139\u013b\u013d\u013f\u0141\u0143\u0145\u0147\u014a\u014c\u014e\u0150\u0152\u0154\u0156\u0158\u015a\u015c\u015e\u0160\u0162\u0164\u0166\u0168\u016a\u016c\u016e\u0170\u0172\u0174\u0176\u0178\u0179\u017b\u017d\u0181\u0182\u0184\u0186\u0187\u0189\u018a\u018b\u018e\u018f\u0190\u0191\u0193\u0194\u0196\u0197\u0198\u019c\u019d\u019f\u01a0\u01a2\u01a4\u01a6\u01a7\u01a9\u01ac\u01ae\u01af\u01b1\u01b2\u01b3\u01b5\u01b7\u01b8\u01bc\u01c4\u01c7\u01ca\u01cd\u01cf\u01d1\u01d3\u01d5\u01d7\u01d9\u01db\u01de\u01e0\u01e2\u01e4\u01e6\u01e8\u01ea\u01ec\u01ee\u01f1\u01f4\u01f6\u01f7\u01f8\u01fa\u01fc\u01fe\u0200\u0202\u0204\u0206\u0208\u020a\u020c\u020e\u0210\u0212\u0214\u0216\u0218\u021a\u021c\u021e\u0220\u0222\u0224\u0226\u0228\u022a\u022c\u022e\u0230\u0232\u023a\u023b\u023d\u023e\u0241\u0243\u0244\u0245\u0246\u0248\u024a\u024c\u024e\u0370\u0372\u0376\u0386\u0388\u0389\u038a\u038c\u038e\u038f\u0391\u0392\u0393\u0394\u0395\u0396\u0397\u0398\u0399\u039a\u039b\u039c\u039d\u039e\u039f\u03a0\u03a1\u03a3\u03a4\u03a5\u03a6\u03a7\u03a8\u03a9\u03aa\u03ab\u03cf\u03d2\u03d3\u03d4\u03d8\u03da\u03dc\u03de\u03e0\u03e2\u03e4\u03e6\u03e8\u03ea\u03ec\u03ee\u03f4\u03f7\u03f9\u03fa\u03fd\u03fe\u03ff\u0400\u0401\u0402\u0403\u0404\u0405\u0406\u0407\u0408\u0409\u040a\u040b\u040c\u040d\u040e\u040f\u0410\u0411\u0412\u0413\u0414\u0415\u0416\u0417\u0418\u0419\u041a\u041b\u041c\u041d\u041e\u041f\u0420\u0421\u0422\u0423\u0424\u0425\u0426\u0427\u0428\u0429\u042a\u042b\u042c\u042d\u042e\u042f\u0460\u0462\u0464\u0466\u0468\u046a\u046c\u046e\u0470\u0472\u0474\u0476\u0478\u047a\u047c\u047e\u0480\u048a\u048c\u048e\u0490\u0492\u0494\u0496\u0498\u049a\u049c\u049e\u04a0\u04a2\u04a4\u04a6\u04a8\u04aa\u04ac\u04ae\u04b0\u04b2\u04b4\u04b6\u04b8\u04ba\u04bc\u04be\u04c0\u04c1\u04c3\u04c5\u04c7\u04c9\u04cb\u04cd\u04d0\u04d2\u04d4\u04d6\u04d8\u04da\u04dc\u04de\u04e0\u04e2\u04e4\u04e6\u04e8\u04ea\u04ec\u04ee\u04f0\u04f2\u04f4\u04f6\u04f8\u04fa\u04fc\u04fe\u0500\u0502\u0504\u0506\u0508\u050a\u050c\u050e\u0510\u0512\u0514\u0516\u0518\u051a\u051c\u051e\u0520\u0522\u0524\u0531\u0532\u0533\u0534\u0535\u0536\u0537\u0538\u0539\u053a\u053b\u053c\u053d\u053e\u053f\u0540\u0541\u0542\u0543\u0544\u0545\u0546\u0547\u0548\u0549\u054a\u054b\u054c\u054d\u054e\u054f\u0550\u0551\u0552\u0553\u0554\u0555\u0556\u10a0\u10a1\u10a2\u10a3\u10a4\u10a5\u10a6\u10a7\u10a8\u10a9\u10aa\u10ab\u10ac\u10ad\u10ae\u10af\u10b0\u10b1\u10b2\u10b3\u10b4\u10b5\u10b6\u10b7\u10b8\u10b9\u10ba\u10bb\u10bc\u10bd\u10be\u10bf\u10c0\u10c1\u10c2\u10c3\u10c4\u10c5\u1e00\u1e02\u1e04\u1e06\u1e08\u1e0a\u1e0c\u1e0e\u1e10\u1e12\u1e14\u1e16\u1e18\u1e1a\u1e1c\u1e1e\u1e20\u1e22\u1e24\u1e26\u1e28\u1e2a\u1e2c\u1e2e\u1e30\u1e32\u1e34\u1e36\u1e38\u1e3a\u1e3c\u1e3e\u1e40\u1e42\u1e44\u1e46\u1e48\u1e4a\u1e4c\u1e4e\u1e50\u1e52\u1e54\u1e56\u1e58\u1e5a\u1e5c\u1e5e\u1e60\u1e62\u1e64\u1e66\u1e68\u1e6a\u1e6c\u1e6e\u1e70\u1e72\u1e74\u1e76\u1e78\u1e7a\u1e7c\u1e7e\u1e80\u1e82\u1e84\u1e86\u1e88\u1e8a\u1e8c\u1e8e\u1e90\u1e92\u1e94\u1e9e\u1ea0\u1ea2\u1ea4\u1ea6\u1ea8\u1eaa\u1eac\u1eae\u1eb0\u1eb2\u1eb4\u1eb6\u1eb8\u1eba\u1ebc\u1ebe\u1ec0\u1ec2\u1ec4\u1ec6\u1ec8\u1eca\u1ecc\u1ece\u1ed0\u1ed2\u1ed4\u1ed6\u1ed8\u1eda\u1edc\u1ede\u1ee0\u1ee2\u1ee4\u1ee6\u1ee8\u1eea\u1eec\u1eee\u1ef0\u1ef2\u1ef4\u1ef6\u1ef8\u1efa\u1efc\u1efe\u1f08\u1f09\u1f0a\u1f0b\u1f0c\u1f0d\u1f0e\u1f0f\u1f18\u1f19\u1f1a\u1f1b\u1f1c\u1f1d\u1f28\u1f29\u1f2a\u1f2b\u1f2c\u1f2d\u1f2e\u1f2f\u1f38\u1f39\u1f3a\u1f3b\u1f3c\u1f3d\u1f3e\u1f3f\u1f48\u1f49\u1f4a\u1f4b\u1f4c\u1f4d\u1f59\u1f5b\u1f5d\u1f5f\u1f68\u1f69\u1f6a\u1f6b\u1f6c\u1f6d\u1f6e\u1f6f\u1fb8\u1fb9\u1fba\u1fbb\u1fc8\u1fc9\u1fca\u1fcb\u1fd8\u1fd9\u1fda\u1fdb\u1fe8\u1fe9\u1fea\u1feb\u1fec\u1ff8\u1ff9\u1ffa\u1ffb\u2102\u2107\u210b\u210c\u210d\u2110\u2111\u2112\u2115\u2119\u211a\u211b\u211c\u211d\u2124\u2126\u2128\u212a\u212b\u212c\
|
2013-09-22 20:39:16 +00:00
|
|
|
|
2013-09-22 21:48:55 +00:00
|
|
|
Mc = '\u0903\u093e\u093f\u0940\u0949\u094a\u094b\u094c\u094e\u0982\u0983\u09be\u09bf\u09c0\u09c7\u09c8\u09cb\u09cc\u09d7\u0a03\u0a3e\u0a3f\u0a40\u0a83\u0abe\u0abf\u0ac0\u0ac9\u0acb\u0acc\u0b02\u0b03\u0b3e\u0b40\u0b47\u0b48\u0b4b\u0b4c\u0b57\u0bbe\u0bbf\u0bc1\u0bc2\u0bc6\u0bc7\u0bc8\u0bca\u0bcb\u0bcc\u0bd7\u0c01\u0c02\u0c03\u0c41\u0c42\u0c43\u0c44\u0c82\u0c83\u0cbe\u0cc0\u0cc1\u0cc2\u0cc3\u0cc4\u0cc7\u0cc8\u0cca\u0ccb\u0cd5\u0cd6\u0d02\u0d03\u0d3e\u0d3f\u0d40\u0d46\u0d47\u0d48\u0d4a\u0d4b\u0d4c\u0d57\u0d82\u0d83\u0dcf\u0dd0\u0dd1\u0dd8\u0dd9\u0dda\u0ddb\u0ddc\u0ddd\u0dde\u0ddf\u0df2\u0df3\u0f3e\u0f3f\u0f7f\u102b\u102c\u1031\u1038\u103b\u103c\u1056\u1057\u1062\u1063\u1064\u1067\u1068\u1069\u106a\u106b\u106c\u106d\u1083\u1084\u1087\u1088\u1089\u108a\u108b\u108c\u108f\u109a\u109b\u109c\u17b6\u17be\u17bf\u17c0\u17c1\u17c2\u17c3\u17c4\u17c5\u17c7\u17c8\u1923\u1924\u1925\u1926\u1929\u192a\u192b\u1930\u1931\u1933\u1934\u1935\u1936\u1937\u1938\u19b0\u19b1\u19b2\u19b3\u19b4\u19b5\u19b6\u19b7\u19b8\u19b9\u19ba\u19bb\u19bc\u19bd\u19be\u19bf\u19c0\u19c8\u19c9\u1a19\u1a1a\u1a1b\u1a55\u1a57\u1a61\u1a63\u1a64\u1a6d\u1a6e\u1a6f\u1a70\u1a71\u1a72\u1b04\u1b35\u1b3b\u1b3d\u1b3e\u1b3f\u1b40\u1b41\u1b43\u1b44\u1b82\u1ba1\u1ba6\u1ba7\u1baa\u1c24\u1c25\u1c26\u1c27\u1c28\u1c29\u1c2a\u1c2b\u1c34\u1c35\u1ce1\u1cf2\ua823\ua824\ua827\ua880\ua881\ua8b4\ua8b5\ua8b6\ua8b7\ua8b8\ua8b9\ua8ba\ua8bb\ua8bc\ua8bd\ua8be\ua8bf\ua8c0\ua8c1\ua8c2\ua8c3\ua952\ua953\ua983\ua9b4\ua9b5\ua9ba\ua9bb\ua9bd\ua9be\ua9bf\ua9c0\uaa2f\uaa30\uaa33\uaa34\uaa4d\uaa7b\uabe3\uabe4\uabe6\uabe7\uabe9\uabea\uabec'
|
2013-09-22 20:39:16 +00:00
|
|
|
|
2013-09-22 21:48:55 +00:00
|
|
|
Me = '\u0488\u0489\u06de\u20dd\u20de\u20df\u20e0\u20e2\u20e3\u20e4\ua670\ua671\ua672'
|
2013-09-22 20:39:16 +00:00
|
|
|
|
2013-09-22 21:48:55 +00:00
|
|
|
Mn = '\u0300\u0301\u0302\u0303\u0304\u0305\u0306\u0307\u0308\u0309\u030a\u030b\u030c\u030d\u030e\u030f\u0310\u0311\u0312\u0313\u0314\u0315\u0316\u0317\u0318\u0319\u031a\u031b\u031c\u031d\u031e\u031f\u0320\u0321\u0322\u0323\u0324\u0325\u0326\u0327\u0328\u0329\u032a\u032b\u032c\u032d\u032e\u032f\u0330\u0331\u0332\u0333\u0334\u0335\u0336\u0337\u0338\u0339\u033a\u033b\u033c\u033d\u033e\u033f\u0340\u0341\u0342\u0343\u0344\u0345\u0346\u0347\u0348\u0349\u034a\u034b\u034c\u034d\u034e\u034f\u0350\u0351\u0352\u0353\u0354\u0355\u0356\u0357\u0358\u0359\u035a\u035b\u035c\u035d\u035e\u035f\u0360\u0361\u0362\u0363\u0364\u0365\u0366\u0367\u0368\u0369\u036a\u036b\u036c\u036d\u036e\u036f\u0483\u0484\u0485\u0486\u0487\u0591\u0592\u0593\u0594\u0595\u0596\u0597\u0598\u0599\u059a\u059b\u059c\u059d\u059e\u059f\u05a0\u05a1\u05a2\u05a3\u05a4\u05a5\u05a6\u05a7\u05a8\u05a9\u05aa\u05ab\u05ac\u05ad\u05ae\u05af\u05b0\u05b1\u05b2\u05b3\u05b4\u05b5\u05b6\u05b7\u05b8\u05b9\u05ba\u05bb\u05bc\u05bd\u05bf\u05c1\u05c2\u05c4\u05c5\u05c7\u0610\u0611\u0612\u0613\u0614\u0615\u0616\u0617\u0618\u0619\u061a\u064b\u064c\u064d\u064e\u064f\u0650\u0651\u0652\u0653\u0654\u0655\u0656\u0657\u0658\u0659\u065a\u065b\u065c\u065d\u065e\u0670\u06d6\u06d7\u06d8\u06d9\u06da\u06db\u06dc\u06df\u06e0\u06e1\u06e2\u06e3\u06e4\u06e7\u06e8\u06ea\u06eb\u06ec\u06ed\u0711\u0730\u0731\u0732\u0733\u0734\u0735\u0736\u0737\u0738\u0739\u073a\u073b\u073c\u073d\u073e\u073f\u0740\u0741\u0742\u0743\u0744\u0745\u0746\u0747\u0748\u0749\u074a\u07a6\u07a7\u07a8\u07a9\u07aa\u07ab\u07ac\u07ad\u07ae\u07af\u07b0\u07eb\u07ec\u07ed\u07ee\u07ef\u07f0\u07f1\u07f2\u07f3\u0816\u0817\u0818\u0819\u081b\u081c\u081d\u081e\u081f\u0820\u0821\u0822\u0823\u0825\u0826\u0827\u0829\u082a\u082b\u082c\u082d\u0900\u0901\u0902\u093c\u0941\u0942\u0943\u0944\u0945\u0946\u0947\u0948\u094d\u0951\u0952\u0953\u0954\u0955\u0962\u0963\u0981\u09bc\u09c1\u09c2\u09c3\u09c4\u09cd\u09e2\u09e3\u0a01\u0a02\u0a3c\u0a41\u0a42\u0a47\u0a48\u0a4b\u0a4c\u0a4d\u0a51\u0a70\u0a71\u0a75\u0a81\u0a82\u0abc\u0ac1\u0ac2\u0ac3\u0ac4\u0ac5\u0ac7\u0ac8\u0acd\u0ae2\u0ae3\u0b01\u0b3c\u0b3f\u0b41\u0b42\u0b43\u0b44\u0b4d\u0b56\u0b62\u0b63\u0b82\u0bc0\u0bcd\u0c3e\u0c3f\u0c40\u0c46\u0c47\u0c48\u0c4a\u0c4b\u0c4c\u0c4d\u0c55\u0c56\u0c62\u0c63\u0cbc\u0cbf\u0cc6\u0ccc\u0ccd\u0ce2\u0ce3\u0d41\u0d42\u0d43\u0d44\u0d4d\u0d62\u0d63\u0dca\u0dd2\u0dd3\u0dd4\u0dd6\u0e31\u0e34\u0e35\u0e36\u0e37\u0e38\u0e39\u0e3a\u0e47\u0e48\u0e49\u0e4a\u0e4b\u0e4c\u0e4d\u0e4e\u0eb1\u0eb4\u0eb5\u0eb6\u0eb7\u0eb8\u0eb9\u0ebb\u0ebc\u0ec8\u0ec9\u0eca\u0ecb\u0ecc\u0ecd\u0f18\u0f19\u0f35\u0f37\u0f39\u0f71\u0f72\u0f73\u0f74\u0f75\u0f76\u0f77\u0f78\u0f79\u0f7a\u0f7b\u0f7c\u0f7d\u0f7e\u0f80\u0f81\u0f82\u0f83\u0f84\u0f86\u0f87\u0f90\u0f91\u0f92\u0f93\u0f94\u0f95\u0f96\u0f97\u0f99\u0f9a\u0f9b\u0f9c\u0f9d\u0f9e\u0f9f\u0fa0\u0fa1\u0fa2\u0fa3\u0fa4\u0fa5\u0fa6\u0fa7\u0fa8\u0fa9\u0faa\u0fab\u0fac\u0fad\u0fae\u0faf\u0fb0\u0fb1\u0fb2\u0fb3\u0fb4\u0fb5\u0fb6\u0fb7\u0fb8\u0fb9\u0fba\u0fbb\u0fbc\u0fc6\u102d\u102e\u102f\u1030\u1032\u1033\u1034\u1035\u1036\u1037\u1039\u103a\u103d\u103e\u1058\u1059\u105e\u105f\u1060\u1071\u1072\u1073\u1074\u1082\u1085\u1086\u108d\u109d\u135f\u1712\u1713\u1714\u1732\u1733\u1734\u1752\u1753\u1772\u1773\u17b7\u17b8\u17b9\u17ba\u17bb\u17bc\u17bd\u17c6\u17c9\u17ca\u17cb\u17cc\u17cd\u17ce\u17cf\u17d0\u17d1\u17d2\u17d3\u17dd\u180b\u180c\u180d\u18a9\u1920\u1921\u1922\u1927\u1928\u1932\u1939\u193a\u193b\u1a17\u1a18\u1a56\u1a58\u1a59\u1a5a\u1a5b\u1a5c\u1a5d\u1a5e\u1a60\u1a62\u1a65\u1a66\u1a67\u1a68\u1a69\u1a6a\u1a6b\u1a6c\u1a73\u1a74\u1a75\u1a76\u1a77\u1a78\u1a79\u1a7a\u1a7b\u1a7c\u1a7f\u1b00\u1b01\u1b02\u1b03\u1b34\u1b36\u1b37\u1b38\u1b39\u1b3a\u1b3c\u1b42\u1b6b\u1b6c\u1b6d\u1b6e\u1b6f\u1b70\u1b71\u1b72\u1b73\u1b80\u1b81\u1ba2\u1ba3\u1ba4\u1ba5\u1ba8\u1ba9\u1c2c\u1c2d\u1c2e\u1c2f\u1c30\u1c31\u1c32\u1c33\u1c36\u1c37\u1cd0\u1cd1\u1cd2\u1cd4\u1cd5\u1cd6\u1cd7\u1cd8\u1cd9\u1cda\u1cdb\u1cdc\u1cdd\u1cde\u1cdf\u1ce0\u1ce2\u1ce3\u1ce4\u1ce5\u1ce6\u1ce7\u1ce8\u1ced\u1dc0\u1dc1\u1dc2\u1dc3\u1dc4\u1dc5\u1dc6\u1dc7\u1dc8\u1dc9\u1dca\u1dcb\u1dcc\u1dcd\u1dce\u1dcf\u1dd0\u1dd1\u1dd2\u1dd3\u1dd4\u1dd5\u1
|
2013-09-22 20:39:16 +00:00
|
|
|
|
2013-09-22 21:48:55 +00:00
|
|
|
Nd = '0123456789\u0660\u0661\u0662\u0663\u0664\u0665\u0666\u0667\u0668\u0669\u06f0\u06f1\u06f2\u06f3\u06f4\u06f5\u06f6\u06f7\u06f8\u06f9\u07c0\u07c1\u07c2\u07c3\u07c4\u07c5\u07c6\u07c7\u07c8\u07c9\u0966\u0967\u0968\u0969\u096a\u096b\u096c\u096d\u096e\u096f\u09e6\u09e7\u09e8\u09e9\u09ea\u09eb\u09ec\u09ed\u09ee\u09ef\u0a66\u0a67\u0a68\u0a69\u0a6a\u0a6b\u0a6c\u0a6d\u0a6e\u0a6f\u0ae6\u0ae7\u0ae8\u0ae9\u0aea\u0aeb\u0aec\u0aed\u0aee\u0aef\u0b66\u0b67\u0b68\u0b69\u0b6a\u0b6b\u0b6c\u0b6d\u0b6e\u0b6f\u0be6\u0be7\u0be8\u0be9\u0bea\u0beb\u0bec\u0bed\u0bee\u0bef\u0c66\u0c67\u0c68\u0c69\u0c6a\u0c6b\u0c6c\u0c6d\u0c6e\u0c6f\u0ce6\u0ce7\u0ce8\u0ce9\u0cea\u0ceb\u0cec\u0ced\u0cee\u0cef\u0d66\u0d67\u0d68\u0d69\u0d6a\u0d6b\u0d6c\u0d6d\u0d6e\u0d6f\u0e50\u0e51\u0e52\u0e53\u0e54\u0e55\u0e56\u0e57\u0e58\u0e59\u0ed0\u0ed1\u0ed2\u0ed3\u0ed4\u0ed5\u0ed6\u0ed7\u0ed8\u0ed9\u0f20\u0f21\u0f22\u0f23\u0f24\u0f25\u0f26\u0f27\u0f28\u0f29\u1040\u1041\u1042\u1043\u1044\u1045\u1046\u1047\u1048\u1049\u1090\u1091\u1092\u1093\u1094\u1095\u1096\u1097\u1098\u1099\u17e0\u17e1\u17e2\u17e3\u17e4\u17e5\u17e6\u17e7\u17e8\u17e9\u1810\u1811\u1812\u1813\u1814\u1815\u1816\u1817\u1818\u1819\u1946\u1947\u1948\u1949\u194a\u194b\u194c\u194d\u194e\u194f\u19d0\u19d1\u19d2\u19d3\u19d4\u19d5\u19d6\u19d7\u19d8\u19d9\u19da\u1a80\u1a81\u1a82\u1a83\u1a84\u1a85\u1a86\u1a87\u1a88\u1a89\u1a90\u1a91\u1a92\u1a93\u1a94\u1a95\u1a96\u1a97\u1a98\u1a99\u1b50\u1b51\u1b52\u1b53\u1b54\u1b55\u1b56\u1b57\u1b58\u1b59\u1bb0\u1bb1\u1bb2\u1bb3\u1bb4\u1bb5\u1bb6\u1bb7\u1bb8\u1bb9\u1c40\u1c41\u1c42\u1c43\u1c44\u1c45\u1c46\u1c47\u1c48\u1c49\u1c50\u1c51\u1c52\u1c53\u1c54\u1c55\u1c56\u1c57\u1c58\u1c59\ua620\ua621\ua622\ua623\ua624\ua625\ua626\ua627\ua628\ua629\ua8d0\ua8d1\ua8d2\ua8d3\ua8d4\ua8d5\ua8d6\ua8d7\ua8d8\ua8d9\ua900\ua901\ua902\ua903\ua904\ua905\ua906\ua907\ua908\ua909\ua9d0\ua9d1\ua9d2\ua9d3\ua9d4\ua9d5\ua9d6\ua9d7\ua9d8\ua9d9\uaa50\uaa51\uaa52\uaa53\uaa54\uaa55\uaa56\uaa57\uaa58\uaa59\uabf0\uabf1\uabf2\uabf3\uabf4\uabf5\uabf6\uabf7\uabf8\uabf9\uff10\uff11\uff12\uff13\uff14\uff15\uff16\uff17\uff18\uff19'
|
2013-09-22 20:39:16 +00:00
|
|
|
|
2013-09-22 21:48:55 +00:00
|
|
|
Nl = '\u16ee\u16ef\u16f0\u2160\u2161\u2162\u2163\u2164\u2165\u2166\u2167\u2168\u2169\u216a\u216b\u216c\u216d\u216e\u216f\u2170\u2171\u2172\u2173\u2174\u2175\u2176\u2177\u2178\u2179\u217a\u217b\u217c\u217d\u217e\u217f\u2180\u2181\u2182\u2185\u2186\u2187\u2188\u3007\u3021\u3022\u3023\u3024\u3025\u3026\u3027\u3028\u3029\u3038\u3039\u303a\ua6e6\ua6e7\ua6e8\ua6e9\ua6ea\ua6eb\ua6ec\ua6ed\ua6ee\ua6ef'
|
2013-09-22 20:39:16 +00:00
|
|
|
|
2013-09-22 21:48:55 +00:00
|
|
|
No = '\xb2\xb3\xb9\xbc\xbd\xbe\u09f4\u09f5\u09f6\u09f7\u09f8\u09f9\u0bf0\u0bf1\u0bf2\u0c78\u0c79\u0c7a\u0c7b\u0c7c\u0c7d\u0c7e\u0d70\u0d71\u0d72\u0d73\u0d74\u0d75\u0f2a\u0f2b\u0f2c\u0f2d\u0f2e\u0f2f\u0f30\u0f31\u0f32\u0f33\u1369\u136a\u136b\u136c\u136d\u136e\u136f\u1370\u1371\u1372\u1373\u1374\u1375\u1376\u1377\u1378\u1379\u137a\u137b\u137c\u17f0\u17f1\u17f2\u17f3\u17f4\u17f5\u17f6\u17f7\u17f8\u17f9\u2070\u2074\u2075\u2076\u2077\u2078\u2079\u2080\u2081\u2082\u2083\u2084\u2085\u2086\u2087\u2088\u2089\u2150\u2151\u2152\u2153\u2154\u2155\u2156\u2157\u2158\u2159\u215a\u215b\u215c\u215d\u215e\u215f\u2189\u2460\u2461\u2462\u2463\u2464\u2465\u2466\u2467\u2468\u2469\u246a\u246b\u246c\u246d\u246e\u246f\u2470\u2471\u2472\u2473\u2474\u2475\u2476\u2477\u2478\u2479\u247a\u247b\u247c\u247d\u247e\u247f\u2480\u2481\u2482\u2483\u2484\u2485\u2486\u2487\u2488\u2489\u248a\u248b\u248c\u248d\u248e\u248f\u2490\u2491\u2492\u2493\u2494\u2495\u2496\u2497\u2498\u2499\u249a\u249b\u24ea\u24eb\u24ec\u24ed\u24ee\u24ef\u24f0\u24f1\u24f2\u24f3\u24f4\u24f5\u24f6\u24f7\u24f8\u24f9\u24fa\u24fb\u24fc\u24fd\u24fe\u24ff\u2776\u2777\u2778\u2779\u277a\u277b\u277c\u277d\u277e\u277f\u2780\u2781\u2782\u2783\u2784\u2785\u2786\u2787\u2788\u2789\u278a\u278b\u278c\u278d\u278e\u278f\u2790\u2791\u2792\u2793\u2cfd\u3192\u3193\u3194\u3195\u3220\u3221\u3222\u3223\u3224\u3225\u3226\u3227\u3228\u3229\u3251\u3252\u3253\u3254\u3255\u3256\u3257\u3258\u3259\u325a\u325b\u325c\u325d\u325e\u325f\u3280\u3281\u3282\u3283\u3284\u3285\u3286\u3287\u3288\u3289\u32b1\u32b2\u32b3\u32b4\u32b5\u32b6\u32b7\u32b8\u32b9\u32ba\u32bb\u32bc\u32bd\u32be\u32bf\ua830\ua831\ua832\ua833\ua834\ua835'
|
2013-09-22 20:39:16 +00:00
|
|
|
|
2013-09-22 21:48:55 +00:00
|
|
|
Pc = '_\u203f\u2040\u2054\ufe33\ufe34\ufe4d\ufe4e\ufe4f\uff3f'
|
2013-09-22 20:39:16 +00:00
|
|
|
|
2013-09-22 21:48:55 +00:00
|
|
|
Pd = '\\-\u058a\u05be\u1400\u1806\u2010\u2011\u2012\u2013\u2014\u2015\u2e17\u2e1a\u301c\u3030\u30a0\ufe31\ufe32\ufe58\ufe63\uff0d'
|
2013-09-22 20:39:16 +00:00
|
|
|
|
2013-09-22 21:48:55 +00:00
|
|
|
Pe = ')]}\u0f3b\u0f3d\u169c\u2046\u207e\u208e\u232a\u2769\u276b\u276d\u276f\u2771\u2773\u2775\u27c6\u27e7\u27e9\u27eb\u27ed\u27ef\u2984\u2986\u2988\u298a\u298c\u298e\u2990\u2992\u2994\u2996\u2998\u29d9\u29db\u29fd\u2e23\u2e25\u2e27\u2e29\u3009\u300b\u300d\u300f\u3011\u3015\u3017\u3019\u301b\u301e\u301f\ufd3f\ufe18\ufe36\ufe38\ufe3a\ufe3c\ufe3e\ufe40\ufe42\ufe44\ufe48\ufe5a\ufe5c\ufe5e\uff09\uff3d\uff5d\uff60\uff63'
|
2013-09-22 20:39:16 +00:00
|
|
|
|
2013-09-22 21:48:55 +00:00
|
|
|
Pf = '\xbb\u2019\u201d\u203a\u2e03\u2e05\u2e0a\u2e0d\u2e1d\u2e21'
|
2013-09-22 20:39:16 +00:00
|
|
|
|
2013-09-22 21:48:55 +00:00
|
|
|
Pi = '\xab\u2018\u201b\u201c\u201f\u2039\u2e02\u2e04\u2e09\u2e0c\u2e1c\u2e20'
|
2013-09-22 20:39:16 +00:00
|
|
|
|
2013-09-22 21:48:55 +00:00
|
|
|
Po = '!"#%&\'*,./:;?@\\\\\xa1\xb7\xbf\u037e\u0387\u055a\u055b\u055c\u055d\u055e\u055f\u0589\u05c0\u05c3\u05c6\u05f3\u05f4\u0609\u060a\u060c\u060d\u061b\u061e\u061f\u066a\u066b\u066c\u066d\u06d4\u0700\u0701\u0702\u0703\u0704\u0705\u0706\u0707\u0708\u0709\u070a\u070b\u070c\u070d\u07f7\u07f8\u07f9\u0830\u0831\u0832\u0833\u0834\u0835\u0836\u0837\u0838\u0839\u083a\u083b\u083c\u083d\u083e\u0964\u0965\u0970\u0df4\u0e4f\u0e5a\u0e5b\u0f04\u0f05\u0f06\u0f07\u0f08\u0f09\u0f0a\u0f0b\u0f0c\u0f0d\u0f0e\u0f0f\u0f10\u0f11\u0f12\u0f85\u0fd0\u0fd1\u0fd2\u0fd3\u0fd4\u104a\u104b\u104c\u104d\u104e\u104f\u10fb\u1361\u1362\u1363\u1364\u1365\u1366\u1367\u1368\u166d\u166e\u16eb\u16ec\u16ed\u1735\u1736\u17d4\u17d5\u17d6\u17d8\u17d9\u17da\u1800\u1801\u1802\u1803\u1804\u1805\u1807\u1808\u1809\u180a\u1944\u1945\u19de\u19df\u1a1e\u1a1f\u1aa0\u1aa1\u1aa2\u1aa3\u1aa4\u1aa5\u1aa6\u1aa8\u1aa9\u1aaa\u1aab\u1aac\u1aad\u1b5a\u1b5b\u1b5c\u1b5d\u1b5e\u1b5f\u1b60\u1c3b\u1c3c\u1c3d\u1c3e\u1c3f\u1c7e\u1c7f\u1cd3\u2016\u2017\u2020\u2021\u2022\u2023\u2024\u2025\u2026\u2027\u2030\u2031\u2032\u2033\u2034\u2035\u2036\u2037\u2038\u203b\u203c\u203d\u203e\u2041\u2042\u2043\u2047\u2048\u2049\u204a\u204b\u204c\u204d\u204e\u204f\u2050\u2051\u2053\u2055\u2056\u2057\u2058\u2059\u205a\u205b\u205c\u205d\u205e\u2cf9\u2cfa\u2cfb\u2cfc\u2cfe\u2cff\u2e00\u2e01\u2e06\u2e07\u2e08\u2e0b\u2e0e\u2e0f\u2e10\u2e11\u2e12\u2e13\u2e14\u2e15\u2e16\u2e18\u2e19\u2e1b\u2e1e\u2e1f\u2e2a\u2e2b\u2e2c\u2e2d\u2e2e\u2e30\u2e31\u3001\u3002\u3003\u303d\u30fb\ua4fe\ua4ff\ua60d\ua60e\ua60f\ua673\ua67e\ua6f2\ua6f3\ua6f4\ua6f5\ua6f6\ua6f7\ua874\ua875\ua876\ua877\ua8ce\ua8cf\ua8f8\ua8f9\ua8fa\ua92e\ua92f\ua95f\ua9c1\ua9c2\ua9c3\ua9c4\ua9c5\ua9c6\ua9c7\ua9c8\ua9c9\ua9ca\ua9cb\ua9cc\ua9cd\ua9de\ua9df\uaa5c\uaa5d\uaa5e\uaa5f\uaade\uaadf\uabeb\ufe10\ufe11\ufe12\ufe13\ufe14\ufe15\ufe16\ufe19\ufe30\ufe45\ufe46\ufe49\ufe4a\ufe4b\ufe4c\ufe50\ufe51\ufe52\ufe54\ufe55\ufe56\ufe57\ufe5f\ufe60\ufe61\ufe68\ufe6a\ufe6b\uff01\uff02\uff03\uff05\uff06\uff07\uff0a\uff0c\uff0e\uff0f\uff1a\uff1b\uff1f\uff20\uff3c\uff61\uff64\uff65'
|
2013-09-22 20:39:16 +00:00
|
|
|
|
2013-09-22 21:48:55 +00:00
|
|
|
Ps = '([{\u0f3a\u0f3c\u169b\u201a\u201e\u2045\u207d\u208d\u2329\u2768\u276a\u276c\u276e\u2770\u2772\u2774\u27c5\u27e6\u27e8\u27ea\u27ec\u27ee\u2983\u2985\u2987\u2989\u298b\u298d\u298f\u2991\u2993\u2995\u2997\u29d8\u29da\u29fc\u2e22\u2e24\u2e26\u2e28\u3008\u300a\u300c\u300e\u3010\u3014\u3016\u3018\u301a\u301d\ufd3e\ufe17\ufe35\ufe37\ufe39\ufe3b\ufe3d\ufe3f\ufe41\ufe43\ufe47\ufe59\ufe5b\ufe5d\uff08\uff3b\uff5b\uff5f\uff62'
|
2013-09-22 20:39:16 +00:00
|
|
|
|
2013-09-22 21:48:55 +00:00
|
|
|
Sc = '$\xa2\xa3\xa4\xa5\u060b\u09f2\u09f3\u09fb\u0af1\u0bf9\u0e3f\u17db\u20a0\u20a1\u20a2\u20a3\u20a4\u20a5\u20a6\u20a7\u20a8\u20a9\u20aa\u20ab\u20ac\u20ad\u20ae\u20af\u20b0\u20b1\u20b2\u20b3\u20b4\u20b5\u20b6\u20b7\u20b8\ua838\ufdfc\ufe69\uff04\uffe0\uffe1\uffe5\uffe6'
|
2013-09-22 20:39:16 +00:00
|
|
|
|
2013-09-22 21:48:55 +00:00
|
|
|
Sk = '^`\xa8\xaf\xb4\xb8\u02c2\u02c3\u02c4\u02c5\u02d2\u02d3\u02d4\u02d5\u02d6\u02d7\u02d8\u02d9\u02da\u02db\u02dc\u02dd\u02de\u02df\u02e5\u02e6\u02e7\u02e8\u02e9\u02ea\u02eb\u02ed\u02ef\u02f0\u02f1\u02f2\u02f3\u02f4\u02f5\u02f6\u02f7\u02f8\u02f9\u02fa\u02fb\u02fc\u02fd\u02fe\u02ff\u0375\u0384\u0385\u1fbd\u1fbf\u1fc0\u1fc1\u1fcd\u1fce\u1fcf\u1fdd\u1fde\u1fdf\u1fed\u1fee\u1fef\u1ffd\u1ffe\u309b\u309c\ua700\ua701\ua702\ua703\ua704\ua705\ua706\ua707\ua708\ua709\ua70a\ua70b\ua70c\ua70d\ua70e\ua70f\ua710\ua711\ua712\ua713\ua714\ua715\ua716\ua720\ua721\ua789\ua78a\uff3e\uff40\uffe3'
|
2013-09-22 20:39:16 +00:00
|
|
|
|
2013-09-22 21:48:55 +00:00
|
|
|
Sm = '+<=>|~\xac\xb1\xd7\xf7\u03f6\u0606\u0607\u0608\u2044\u2052\u207a\u207b\u207c\u208a\u208b\u208c\u2140\u2141\u2142\u2143\u2144\u214b\u2190\u2191\u2192\u2193\u2194\u219a\u219b\u21a0\u21a3\u21a6\u21ae\u21ce\u21cf\u21d2\u21d4\u21f4\u21f5\u21f6\u21f7\u21f8\u21f9\u21fa\u21fb\u21fc\u21fd\u21fe\u21ff\u2200\u2201\u2202\u2203\u2204\u2205\u2206\u2207\u2208\u2209\u220a\u220b\u220c\u220d\u220e\u220f\u2210\u2211\u2212\u2213\u2214\u2215\u2216\u2217\u2218\u2219\u221a\u221b\u221c\u221d\u221e\u221f\u2220\u2221\u2222\u2223\u2224\u2225\u2226\u2227\u2228\u2229\u222a\u222b\u222c\u222d\u222e\u222f\u2230\u2231\u2232\u2233\u2234\u2235\u2236\u2237\u2238\u2239\u223a\u223b\u223c\u223d\u223e\u223f\u2240\u2241\u2242\u2243\u2244\u2245\u2246\u2247\u2248\u2249\u224a\u224b\u224c\u224d\u224e\u224f\u2250\u2251\u2252\u2253\u2254\u2255\u2256\u2257\u2258\u2259\u225a\u225b\u225c\u225d\u225e\u225f\u2260\u2261\u2262\u2263\u2264\u2265\u2266\u2267\u2268\u2269\u226a\u226b\u226c\u226d\u226e\u226f\u2270\u2271\u2272\u2273\u2274\u2275\u2276\u2277\u2278\u2279\u227a\u227b\u227c\u227d\u227e\u227f\u2280\u2281\u2282\u2283\u2284\u2285\u2286\u2287\u2288\u2289\u228a\u228b\u228c\u228d\u228e\u228f\u2290\u2291\u2292\u2293\u2294\u2295\u2296\u2297\u2298\u2299\u229a\u229b\u229c\u229d\u229e\u229f\u22a0\u22a1\u22a2\u22a3\u22a4\u22a5\u22a6\u22a7\u22a8\u22a9\u22aa\u22ab\u22ac\u22ad\u22ae\u22af\u22b0\u22b1\u22b2\u22b3\u22b4\u22b5\u22b6\u22b7\u22b8\u22b9\u22ba\u22bb\u22bc\u22bd\u22be\u22bf\u22c0\u22c1\u22c2\u22c3\u22c4\u22c5\u22c6\u22c7\u22c8\u22c9\u22ca\u22cb\u22cc\u22cd\u22ce\u22cf\u22d0\u22d1\u22d2\u22d3\u22d4\u22d5\u22d6\u22d7\u22d8\u22d9\u22da\u22db\u22dc\u22dd\u22de\u22df\u22e0\u22e1\u22e2\u22e3\u22e4\u22e5\u22e6\u22e7\u22e8\u22e9\u22ea\u22eb\u22ec\u22ed\u22ee\u22ef\u22f0\u22f1\u22f2\u22f3\u22f4\u22f5\u22f6\u22f7\u22f8\u22f9\u22fa\u22fb\u22fc\u22fd\u22fe\u22ff\u2308\u2309\u230a\u230b\u2320\u2321\u237c\u239b\u239c\u239d\u239e\u239f\u23a0\u23a1\u23a2\u23a3\u23a4\u23a5\u23a6\u23a7\u23a8\u23a9\u23aa\u23ab\u23ac\u23ad\u23ae\u23af\u23b0\u23b1\u23b2\u23b3\u23dc\u23dd\u23de\u23df\u23e0\u23e1\u25b7\u25c1\u25f8\u25f9\u25fa\u25fb\u25fc\u25fd\u25fe\u25ff\u266f\u27c0\u27c1\u27c2\u27c3\u27c4\u27c7\u27c8\u27c9\u27ca\u27cc\u27d0\u27d1\u27d2\u27d3\u27d4\u27d5\u27d6\u27d7\u27d8\u27d9\u27da\u27db\u27dc\u27dd\u27de\u27df\u27e0\u27e1\u27e2\u27e3\u27e4\u27e5\u27f0\u27f1\u27f2\u27f3\u27f4\u27f5\u27f6\u27f7\u27f8\u27f9\u27fa\u27fb\u27fc\u27fd\u27fe\u27ff\u2900\u2901\u2902\u2903\u2904\u2905\u2906\u2907\u2908\u2909\u290a\u290b\u290c\u290d\u290e\u290f\u2910\u2911\u2912\u2913\u2914\u2915\u2916\u2917\u2918\u2919\u291a\u291b\u291c\u291d\u291e\u291f\u2920\u2921\u2922\u2923\u2924\u2925\u2926\u2927\u2928\u2929\u292a\u292b\u292c\u292d\u292e\u292f\u2930\u2931\u2932\u2933\u2934\u2935\u2936\u2937\u2938\u2939\u293a\u293b\u293c\u293d\u293e\u293f\u2940\u2941\u2942\u2943\u2944\u2945\u2946\u2947\u2948\u2949\u294a\u294b\u294c\u294d\u294e\u294f\u2950\u2951\u2952\u2953\u2954\u2955\u2956\u2957\u2958\u2959\u295a\u295b\u295c\u295d\u295e\u295f\u2960\u2961\u2962\u2963\u2964\u2965\u2966\u2967\u2968\u2969\u296a\u296b\u296c\u296d\u296e\u296f\u2970\u2971\u2972\u2973\u2974\u2975\u2976\u2977\u2978\u2979\u297a\u297b\u297c\u297d\u297e\u297f\u2980\u2981\u2982\u2999\u299a\u299b\u299c\u299d\u299e\u299f\u29a0\u29a1\u29a2\u29a3\u29a4\u29a5\u29a6\u29a7\u29a8\u29a9\u29aa\u29ab\u29ac\u29ad\u29ae\u29af\u29b0\u29b1\u29b2\u29b3\u29b4\u29b5\u29b6\u29b7\u29b8\u29b9\u29ba\u29bb\u29bc\u29bd\u29be\u29bf\u29c0\u29c1\u29c2\u29c3\u29c4\u29c5\u29c6\u29c7\u29c8\u29c9\u29ca\u29cb\u29cc\u29cd\u29ce\u29cf\u29d0\u29d1\u29d2\u29d3\u29d4\u29d5\u29d6\u29d7\u29dc\u29dd\u29de\u29df\u29e0\u29e1\u29e2\u29e3\u29e4\u29e5\u29e6\u29e7\u29e8\u29e9\u29ea\u29eb\u29ec\u29ed\u29ee\u29ef\u29f0\u29f1\u29f2\u29f3\u29f4\u29f5\u29f6\u29f7\u29f8\u29f9\u29fa\u29fb\u29fe\u29ff\u2a00\u2a01\u2a02\u2a03\u2a04\u2a05\u2a06\u2a07\u2a08\u2a09\u2a0a\u2a0b\u2a0c\u2a0d\u2a0e\u2a0f\u2a10\u2a11\u2a12\u2a13\u2a14\u2a15\u2a16\u2a17\u2a18\u2a19\u2a1a\u2a1b\u2a1c\u2a1d\u2a1e\u2a1f\u2a20\u2a21\u2a22\u2a23\u2a24\u2a25\u2a26\u2a27\u2a28\u2a29\u2a2a\u2a2b\u2a2c\u2a2d\u2a2e\u2a2f\u2a30\u2a31\u2a32\u2a3
|
2013-09-22 20:39:16 +00:00
|
|
|
|
2013-09-22 21:48:55 +00:00
|
|
|
So = '\xa6\xa7\xa9\xae\xb0\xb6\u0482\u060e\u060f\u06e9\u06fd\u06fe\u07f6\u09fa\u0b70\u0bf3\u0bf4\u0bf5\u0bf6\u0bf7\u0bf8\u0bfa\u0c7f\u0cf1\u0cf2\u0d79\u0f01\u0f02\u0f03\u0f13\u0f14\u0f15\u0f16\u0f17\u0f1a\u0f1b\u0f1c\u0f1d\u0f1e\u0f1f\u0f34\u0f36\u0f38\u0fbe\u0fbf\u0fc0\u0fc1\u0fc2\u0fc3\u0fc4\u0fc5\u0fc7\u0fc8\u0fc9\u0fca\u0fcb\u0fcc\u0fce\u0fcf\u0fd5\u0fd6\u0fd7\u0fd8\u109e\u109f\u1360\u1390\u1391\u1392\u1393\u1394\u1395\u1396\u1397\u1398\u1399\u1940\u19e0\u19e1\u19e2\u19e3\u19e4\u19e5\u19e6\u19e7\u19e8\u19e9\u19ea\u19eb\u19ec\u19ed\u19ee\u19ef\u19f0\u19f1\u19f2\u19f3\u19f4\u19f5\u19f6\u19f7\u19f8\u19f9\u19fa\u19fb\u19fc\u19fd\u19fe\u19ff\u1b61\u1b62\u1b63\u1b64\u1b65\u1b66\u1b67\u1b68\u1b69\u1b6a\u1b74\u1b75\u1b76\u1b77\u1b78\u1b79\u1b7a\u1b7b\u1b7c\u2100\u2101\u2103\u2104\u2105\u2106\u2108\u2109\u2114\u2116\u2117\u2118\u211e\u211f\u2120\u2121\u2122\u2123\u2125\u2127\u2129\u212e\u213a\u213b\u214a\u214c\u214d\u214f\u2195\u2196\u2197\u2198\u2199\u219c\u219d\u219e\u219f\u21a1\u21a2\u21a4\u21a5\u21a7\u21a8\u21a9\u21aa\u21ab\u21ac\u21ad\u21af\u21b0\u21b1\u21b2\u21b3\u21b4\u21b5\u21b6\u21b7\u21b8\u21b9\u21ba\u21bb\u21bc\u21bd\u21be\u21bf\u21c0\u21c1\u21c2\u21c3\u21c4\u21c5\u21c6\u21c7\u21c8\u21c9\u21ca\u21cb\u21cc\u21cd\u21d0\u21d1\u21d3\u21d5\u21d6\u21d7\u21d8\u21d9\u21da\u21db\u21dc\u21dd\u21de\u21df\u21e0\u21e1\u21e2\u21e3\u21e4\u21e5\u21e6\u21e7\u21e8\u21e9\u21ea\u21eb\u21ec\u21ed\u21ee\u21ef\u21f0\u21f1\u21f2\u21f3\u2300\u2301\u2302\u2303\u2304\u2305\u2306\u2307\u230c\u230d\u230e\u230f\u2310\u2311\u2312\u2313\u2314\u2315\u2316\u2317\u2318\u2319\u231a\u231b\u231c\u231d\u231e\u231f\u2322\u2323\u2324\u2325\u2326\u2327\u2328\u232b\u232c\u232d\u232e\u232f\u2330\u2331\u2332\u2333\u2334\u2335\u2336\u2337\u2338\u2339\u233a\u233b\u233c\u233d\u233e\u233f\u2340\u2341\u2342\u2343\u2344\u2345\u2346\u2347\u2348\u2349\u234a\u234b\u234c\u234d\u234e\u234f\u2350\u2351\u2352\u2353\u2354\u2355\u2356\u2357\u2358\u2359\u235a\u235b\u235c\u235d\u235e\u235f\u2360\u2361\u2362\u2363\u2364\u2365\u2366\u2367\u2368\u2369\u236a\u236b\u236c\u236d\u236e\u236f\u2370\u2371\u2372\u2373\u2374\u2375\u2376\u2377\u2378\u2379\u237a\u237b\u237d\u237e\u237f\u2380\u2381\u2382\u2383\u2384\u2385\u2386\u2387\u2388\u2389\u238a\u238b\u238c\u238d\u238e\u238f\u2390\u2391\u2392\u2393\u2394\u2395\u2396\u2397\u2398\u2399\u239a\u23b4\u23b5\u23b6\u23b7\u23b8\u23b9\u23ba\u23bb\u23bc\u23bd\u23be\u23bf\u23c0\u23c1\u23c2\u23c3\u23c4\u23c5\u23c6\u23c7\u23c8\u23c9\u23ca\u23cb\u23cc\u23cd\u23ce\u23cf\u23d0\u23d1\u23d2\u23d3\u23d4\u23d5\u23d6\u23d7\u23d8\u23d9\u23da\u23db\u23e2\u23e3\u23e4\u23e5\u23e6\u23e7\u23e8\u2400\u2401\u2402\u2403\u2404\u2405\u2406\u2407\u2408\u2409\u240a\u240b\u240c\u240d\u240e\u240f\u2410\u2411\u2412\u2413\u2414\u2415\u2416\u2417\u2418\u2419\u241a\u241b\u241c\u241d\u241e\u241f\u2420\u2421\u2422\u2423\u2424\u2425\u2426\u2440\u2441\u2442\u2443\u2444\u2445\u2446\u2447\u2448\u2449\u244a\u249c\u249d\u249e\u249f\u24a0\u24a1\u24a2\u24a3\u24a4\u24a5\u24a6\u24a7\u24a8\u24a9\u24aa\u24ab\u24ac\u24ad\u24ae\u24af\u24b0\u24b1\u24b2\u24b3\u24b4\u24b5\u24b6\u24b7\u24b8\u24b9\u24ba\u24bb\u24bc\u24bd\u24be\u24bf\u24c0\u24c1\u24c2\u24c3\u24c4\u24c5\u24c6\u24c7\u24c8\u24c9\u24ca\u24cb\u24cc\u24cd\u24ce\u24cf\u24d0\u24d1\u24d2\u24d3\u24d4\u24d5\u24d6\u24d7\u24d8\u24d9\u24da\u24db\u24dc\u24dd\u24de\u24df\u24e0\u24e1\u24e2\u24e3\u24e4\u24e5\u24e6\u24e7\u24e8\u24e9\u2500\u2501\u2502\u2503\u2504\u2505\u2506\u2507\u2508\u2509\u250a\u250b\u250c\u250d\u250e\u250f\u2510\u2511\u2512\u2513\u2514\u2515\u2516\u2517\u2518\u2519\u251a\u251b\u251c\u251d\u251e\u251f\u2520\u2521\u2522\u2523\u2524\u2525\u2526\u2527\u2528\u2529\u252a\u252b\u252c\u252d\u252e\u252f\u2530\u2531\u2532\u2533\u2534\u2535\u2536\u2537\u2538\u2539\u253a\u253b\u253c\u253d\u253e\u253f\u2540\u2541\u2542\u2543\u2544\u2545\u2546\u2547\u2548\u2549\u254a\u254b\u254c\u254d\u254e\u254f\u2550\u2551\u2552\u2553\u2554\u2555\u2556\u2557\u2558\u2559\u255a\u255b\u255c\u255d\u255e\u255f\u2560\u2561\u2562\u2563\u2564\u2565\u2566\u2567\u2568\u2569\u256a\u256b\u256c\u256d\u256e\u256f\u2570\u2571\u2572\u2573\u2574\u2575\u2576\u2577\u2578\u2
|
2013-09-22 20:39:16 +00:00
|
|
|
|
2013-09-22 21:48:55 +00:00
|
|
|
Zl = '\u2028'
|
2013-09-22 20:39:16 +00:00
|
|
|
|
2013-09-22 21:48:55 +00:00
|
|
|
Zp = '\u2029'
|
2013-09-22 20:39:16 +00:00
|
|
|
|
2013-09-22 21:48:55 +00:00
|
|
|
Zs = ' \xa0\u1680\u180e\u2000\u2001\u2002\u2003\u2004\u2005\u2006\u2007\u2008\u2009\u200a\u202f\u205f\u3000'
|
2013-09-22 20:39:16 +00:00
|
|
|
|
|
|
|
cats = ['Cc', 'Cf', 'Cn', 'Co', 'Cs', 'Ll', 'Lm', 'Lo', 'Lt', 'Lu', 'Mc', 'Me', 'Mn', 'Nd', 'Nl', 'No', 'Pc', 'Pd', 'Pe', 'Pf', 'Pi', 'Po', 'Ps', 'Sc', 'Sk', 'Sm', 'So', 'Zl', 'Zp', 'Zs']
|
|
|
|
|
|
|
|
# Generated from unidata 5.2.0
|
|
|
|
|
|
|
|
def combine(*args):
|
2013-09-22 21:48:55 +00:00
|
|
|
return ''.join([globals()[cat] for cat in args])
|
2013-09-22 20:39:16 +00:00
|
|
|
|
2013-09-22 21:48:55 +00:00
|
|
|
xid_start = '\u0041-\u005A\u005F\u0061-\u007A\u00AA\u00B5\u00BA\u00C0-\u00D6\u00D8-\u00F6\u00F8-\u01BA\u01BB\u01BC-\u01BF\u01C0-\u01C3\u01C4-\u0241\u0250-\u02AF\u02B0-\u02C1\u02C6-\u02D1\u02E0-\u02E4\u02EE\u0386\u0388-\u038A\u038C\u038E-\u03A1\u03A3-\u03CE\u03D0-\u03F5\u03F7-\u0481\u048A-\u04CE\u04D0-\u04F9\u0500-\u050F\u0531-\u0556\u0559\u0561-\u0587\u05D0-\u05EA\u05F0-\u05F2\u0621-\u063A\u0640\u0641-\u064A\u066E-\u066F\u0671-\u06D3\u06D5\u06E5-\u06E6\u06EE-\u06EF\u06FA-\u06FC\u06FF\u0710\u0712-\u072F\u074D-\u076D\u0780-\u07A5\u07B1\u0904-\u0939\u093D\u0950\u0958-\u0961\u097D\u0985-\u098C\u098F-\u0990\u0993-\u09A8\u09AA-\u09B0\u09B2\u09B6-\u09B9\u09BD\u09CE\u09DC-\u09DD\u09DF-\u09E1\u09F0-\u09F1\u0A05-\u0A0A\u0A0F-\u0A10\u0A13-\u0A28\u0A2A-\u0A30\u0A32-\u0A33\u0A35-\u0A36\u0A38-\u0A39\u0A59-\u0A5C\u0A5E\u0A72-\u0A74\u0A85-\u0A8D\u0A8F-\u0A91\u0A93-\u0AA8\u0AAA-\u0AB0\u0AB2-\u0AB3\u0AB5-\u0AB9\u0ABD\u0AD0\u0AE0-\u0AE1\u0B05-\u0B0C\u0B0F-\u0B10\u0B13-\u0B28\u0B2A-\u0B30\u0B32-\u0B33\u0B35-\u0B39\u0B3D\u0B5C-\u0B5D\u0B5F-\u0B61\u0B71\u0B83\u0B85-\u0B8A\u0B8E-\u0B90\u0B92-\u0B95\u0B99-\u0B9A\u0B9C\u0B9E-\u0B9F\u0BA3-\u0BA4\u0BA8-\u0BAA\u0BAE-\u0BB9\u0C05-\u0C0C\u0C0E-\u0C10\u0C12-\u0C28\u0C2A-\u0C33\u0C35-\u0C39\u0C60-\u0C61\u0C85-\u0C8C\u0C8E-\u0C90\u0C92-\u0CA8\u0CAA-\u0CB3\u0CB5-\u0CB9\u0CBD\u0CDE\u0CE0-\u0CE1\u0D05-\u0D0C\u0D0E-\u0D10\u0D12-\u0D28\u0D2A-\u0D39\u0D60-\u0D61\u0D85-\u0D96\u0D9A-\u0DB1\u0DB3-\u0DBB\u0DBD\u0DC0-\u0DC6\u0E01-\u0E30\u0E32\u0E40-\u0E45\u0E46\u0E81-\u0E82\u0E84\u0E87-\u0E88\u0E8A\u0E8D\u0E94-\u0E97\u0E99-\u0E9F\u0EA1-\u0EA3\u0EA5\u0EA7\u0EAA-\u0EAB\u0EAD-\u0EB0\u0EB2\u0EBD\u0EC0-\u0EC4\u0EC6\u0EDC-\u0EDD\u0F00\u0F40-\u0F47\u0F49-\u0F6A\u0F88-\u0F8B\u1000-\u1021\u1023-\u1027\u1029-\u102A\u1050-\u1055\u10A0-\u10C5\u10D0-\u10FA\u10FC\u1100-\u1159\u115F-\u11A2\u11A8-\u11F9\u1200-\u1248\u124A-\u124D\u1250-\u1256\u1258\u125A-\u125D\u1260-\u1288\u128A-\u128D\u1290-\u12B0\u12B2-\u12B5\u12B8-\u12BE\u12C0\u12C2-\u12C5\u12C8-\u12D6\u12D8-\u1310\u1312-\u1315\u1318-\u135A\u1380-\u138F\u13A0-\u13F4\u1401-\u166C\u166F-\u1676\u1681-\u169A\u16A0-\u16EA\u16EE-\u16F0\u1700-\u170C\u170E-\u1711\u1720-\u1731\u1740-\u1751\u1760-\u176C\u176E-\u1770\u1780-\u17B3\u17D7\u17DC\u1820-\u1842\u1843\u1844-\u1877\u1880-\u18A8\u1900-\u191C\u1950-\u196D\u1970-\u1974\u1980-\u19A9\u19C1-\u19C7\u1A00-\u1A16\u1D00-\u1D2B\u1D2C-\u1D61\u1D62-\u1D77\u1D78\u1D79-\u1D9A\u1D9B-\u1DBF\u1E00-\u1E9B\u1EA0-\u1EF9\u1F00-\u1F15\u1F18-\u1F1D\u1F20-\u1F45\u1F48-\u1F4D\u1F50-\u1F57\u1F59\u1F5B\u1F5D\u1F5F-\u1F7D\u1F80-\u1FB4\u1FB6-\u1FBC\u1FBE\u1FC2-\u1FC4\u1FC6-\u1FCC\u1FD0-\u1FD3\u1FD6-\u1FDB\u1FE0-\u1FEC\u1FF2-\u1FF4\u1FF6-\u1FFC\u2071\u207F\u2090-\u2094\u2102\u2107\u210A-\u2113\u2115\u2118\u2119-\u211D\u2124\u2126\u2128\u212A-\u212D\u212E\u212F-\u2131\u2133-\u2134\u2135-\u2138\u2139\u213C-\u213F\u2145-\u2149\u2160-\u2183\u2C00-\u2C2E\u2C30-\u2C5E\u2C80-\u2CE4\u2D00-\u2D25\u2D30-\u2D65\u2D6F\u2D80-\u2D96\u2DA0-\u2DA6\u2DA8-\u2DAE\u2DB0-\u2DB6\u2DB8-\u2DBE\u2DC0-\u2DC6\u2DC8-\u2DCE\u2DD0-\u2DD6\u2DD8-\u2DDE\u3005\u3006\u3007\u3021-\u3029\u3031-\u3035\u3038-\u303A\u303B\u303C\u3041-\u3096\u309D-\u309E\u309F\u30A1-\u30FA\u30FC-\u30FE\u30FF\u3105-\u312C\u3131-\u318E\u31A0-\u31B7\u31F0-\u31FF\u3400-\u4DB5\u4E00-\u9FBB\uA000-\uA014\uA015\uA016-\uA48C\uA800-\uA801\uA803-\uA805\uA807-\uA80A\uA80C-\uA822\uAC00-\uD7A3\uF900-\uFA2D\uFA30-\uFA6A\uFA70-\uFAD9\uFB00-\uFB06\uFB13-\uFB17\uFB1D\uFB1F-\uFB28\uFB2A-\uFB36\uFB38-\uFB3C\uFB3E\uFB40-\uFB41\uFB43-\uFB44\uFB46-\uFBB1\uFBD3-\uFC5D\uFC64-\uFD3D\uFD50-\uFD8F\uFD92-\uFDC7\uFDF0-\uFDF9\uFE71\uFE73\uFE77\uFE79\uFE7B\uFE7D\uFE7F-\uFEFC\uFF21-\uFF3A\uFF41-\uFF5A\uFF66-\uFF6F\uFF70\uFF71-\uFF9D\uFFA0-\uFFBE\uFFC2-\uFFC7\uFFCA-\uFFCF\uFFD2-\uFFD7\uFFDA-\uFFDC'
|
2013-09-22 20:39:16 +00:00
|
|
|
|
2013-09-22 21:48:55 +00:00
|
|
|
xid_continue = '\u0030-\u0039\u0041-\u005A\u005F\u0061-\u007A\u00AA\u00B5\u00B7\u00BA\u00C0-\u00D6\u00D8-\u00F6\u00F8-\u01BA\u01BB\u01BC-\u01BF\u01C0-\u01C3\u01C4-\u0241\u0250-\u02AF\u02B0-\u02C1\u02C6-\u02D1\u02E0-\u02E4\u02EE\u0300-\u036F\u0386\u0388-\u038A\u038C\u038E-\u03A1\u03A3-\u03CE\u03D0-\u03F5\u03F7-\u0481\u0483-\u0486\u048A-\u04CE\u04D0-\u04F9\u0500-\u050F\u0531-\u0556\u0559\u0561-\u0587\u0591-\u05B9\u05BB-\u05BD\u05BF\u05C1-\u05C2\u05C4-\u05C5\u05C7\u05D0-\u05EA\u05F0-\u05F2\u0610-\u0615\u0621-\u063A\u0640\u0641-\u064A\u064B-\u065E\u0660-\u0669\u066E-\u066F\u0670\u0671-\u06D3\u06D5\u06D6-\u06DC\u06DF-\u06E4\u06E5-\u06E6\u06E7-\u06E8\u06EA-\u06ED\u06EE-\u06EF\u06F0-\u06F9\u06FA-\u06FC\u06FF\u0710\u0711\u0712-\u072F\u0730-\u074A\u074D-\u076D\u0780-\u07A5\u07A6-\u07B0\u07B1\u0901-\u0902\u0903\u0904-\u0939\u093C\u093D\u093E-\u0940\u0941-\u0948\u0949-\u094C\u094D\u0950\u0951-\u0954\u0958-\u0961\u0962-\u0963\u0966-\u096F\u097D\u0981\u0982-\u0983\u0985-\u098C\u098F-\u0990\u0993-\u09A8\u09AA-\u09B0\u09B2\u09B6-\u09B9\u09BC\u09BD\u09BE-\u09C0\u09C1-\u09C4\u09C7-\u09C8\u09CB-\u09CC\u09CD\u09CE\u09D7\u09DC-\u09DD\u09DF-\u09E1\u09E2-\u09E3\u09E6-\u09EF\u09F0-\u09F1\u0A01-\u0A02\u0A03\u0A05-\u0A0A\u0A0F-\u0A10\u0A13-\u0A28\u0A2A-\u0A30\u0A32-\u0A33\u0A35-\u0A36\u0A38-\u0A39\u0A3C\u0A3E-\u0A40\u0A41-\u0A42\u0A47-\u0A48\u0A4B-\u0A4D\u0A59-\u0A5C\u0A5E\u0A66-\u0A6F\u0A70-\u0A71\u0A72-\u0A74\u0A81-\u0A82\u0A83\u0A85-\u0A8D\u0A8F-\u0A91\u0A93-\u0AA8\u0AAA-\u0AB0\u0AB2-\u0AB3\u0AB5-\u0AB9\u0ABC\u0ABD\u0ABE-\u0AC0\u0AC1-\u0AC5\u0AC7-\u0AC8\u0AC9\u0ACB-\u0ACC\u0ACD\u0AD0\u0AE0-\u0AE1\u0AE2-\u0AE3\u0AE6-\u0AEF\u0B01\u0B02-\u0B03\u0B05-\u0B0C\u0B0F-\u0B10\u0B13-\u0B28\u0B2A-\u0B30\u0B32-\u0B33\u0B35-\u0B39\u0B3C\u0B3D\u0B3E\u0B3F\u0B40\u0B41-\u0B43\u0B47-\u0B48\u0B4B-\u0B4C\u0B4D\u0B56\u0B57\u0B5C-\u0B5D\u0B5F-\u0B61\u0B66-\u0B6F\u0B71\u0B82\u0B83\u0B85-\u0B8A\u0B8E-\u0B90\u0B92-\u0B95\u0B99-\u0B9A\u0B9C\u0B9E-\u0B9F\u0BA3-\u0BA4\u0BA8-\u0BAA\u0BAE-\u0BB9\u0BBE-\u0BBF\u0BC0\u0BC1-\u0BC2\u0BC6-\u0BC8\u0BCA-\u0BCC\u0BCD\u0BD7\u0BE6-\u0BEF\u0C01-\u0C03\u0C05-\u0C0C\u0C0E-\u0C10\u0C12-\u0C28\u0C2A-\u0C33\u0C35-\u0C39\u0C3E-\u0C40\u0C41-\u0C44\u0C46-\u0C48\u0C4A-\u0C4D\u0C55-\u0C56\u0C60-\u0C61\u0C66-\u0C6F\u0C82-\u0C83\u0C85-\u0C8C\u0C8E-\u0C90\u0C92-\u0CA8\u0CAA-\u0CB3\u0CB5-\u0CB9\u0CBC\u0CBD\u0CBE\u0CBF\u0CC0-\u0CC4\u0CC6\u0CC7-\u0CC8\u0CCA-\u0CCB\u0CCC-\u0CCD\u0CD5-\u0CD6\u0CDE\u0CE0-\u0CE1\u0CE6-\u0CEF\u0D02-\u0D03\u0D05-\u0D0C\u0D0E-\u0D10\u0D12-\u0D28\u0D2A-\u0D39\u0D3E-\u0D40\u0D41-\u0D43\u0D46-\u0D48\u0D4A-\u0D4C\u0D4D\u0D57\u0D60-\u0D61\u0D66-\u0D6F\u0D82-\u0D83\u0D85-\u0D96\u0D9A-\u0DB1\u0DB3-\u0DBB\u0DBD\u0DC0-\u0DC6\u0DCA\u0DCF-\u0DD1\u0DD2-\u0DD4\u0DD6\u0DD8-\u0DDF\u0DF2-\u0DF3\u0E01-\u0E30\u0E31\u0E32-\u0E33\u0E34-\u0E3A\u0E40-\u0E45\u0E46\u0E47-\u0E4E\u0E50-\u0E59\u0E81-\u0E82\u0E84\u0E87-\u0E88\u0E8A\u0E8D\u0E94-\u0E97\u0E99-\u0E9F\u0EA1-\u0EA3\u0EA5\u0EA7\u0EAA-\u0EAB\u0EAD-\u0EB0\u0EB1\u0EB2-\u0EB3\u0EB4-\u0EB9\u0EBB-\u0EBC\u0EBD\u0EC0-\u0EC4\u0EC6\u0EC8-\u0ECD\u0ED0-\u0ED9\u0EDC-\u0EDD\u0F00\u0F18-\u0F19\u0F20-\u0F29\u0F35\u0F37\u0F39\u0F3E-\u0F3F\u0F40-\u0F47\u0F49-\u0F6A\u0F71-\u0F7E\u0F7F\u0F80-\u0F84\u0F86-\u0F87\u0F88-\u0F8B\u0F90-\u0F97\u0F99-\u0FBC\u0FC6\u1000-\u1021\u1023-\u1027\u1029-\u102A\u102C\u102D-\u1030\u1031\u1032\u1036-\u1037\u1038\u1039\u1040-\u1049\u1050-\u1055\u1056-\u1057\u1058-\u1059\u10A0-\u10C5\u10D0-\u10FA\u10FC\u1100-\u1159\u115F-\u11A2\u11A8-\u11F9\u1200-\u1248\u124A-\u124D\u1250-\u1256\u1258\u125A-\u125D\u1260-\u1288\u128A-\u128D\u1290-\u12B0\u12B2-\u12B5\u12B8-\u12BE\u12C0\u12C2-\u12C5\u12C8-\u12D6\u12D8-\u1310\u1312-\u1315\u1318-\u135A\u135F\u1369-\u1371\u1380-\u138F\u13A0-\u13F4\u1401-\u166C\u166F-\u1676\u1681-\u169A\u16A0-\u16EA\u16EE-\u16F0\u1700-\u170C\u170E-\u1711\u1712-\u1714\u1720-\u1731\u1732-\u1734\u1740-\u1751\u1752-\u1753\u1760-\u176C\u176E-\u1770\u1772-\u1773\u1780-\u17B3\u17B6\u17B7-\u17BD\u17BE-\u17C5\u17C6\u17C7-\u17C8\u17C9-\u17D3\u17D7\u17DC\u17DD\u17E0-\u17E9\u180B-\u180D\u1810-\u1819\u1820-\u1842\u1843\u1844-\u1877\u1880-\u18A8\u18A9\u1900-\u191C\u1920-\u1922\u1923-\
|
2013-09-22 20:39:16 +00:00
|
|
|
|
|
|
|
def allexcept(*args):
|
|
|
|
newcats = cats[:]
|
|
|
|
for arg in args:
|
|
|
|
newcats.remove(arg)
|
2013-09-22 21:48:55 +00:00
|
|
|
return ''.join([globals()[cat] for cat in newcats])
|
2013-09-22 20:39:16 +00:00
|
|
|
|
|
|
|
if __name__ == '__main__':
|
|
|
|
import unicodedata
|
|
|
|
|
|
|
|
categories = {}
|
|
|
|
|
|
|
|
f = open(__file__)
|
|
|
|
try:
|
|
|
|
content = f.read()
|
|
|
|
finally:
|
|
|
|
f.close()
|
|
|
|
|
|
|
|
header = content[:content.find('Cc =')]
|
|
|
|
footer = content[content.find("def combine("):]
|
|
|
|
|
|
|
|
for code in range(65535):
|
2013-09-22 21:48:55 +00:00
|
|
|
c = chr(code)
|
2013-09-22 20:39:16 +00:00
|
|
|
cat = unicodedata.category(c)
|
|
|
|
if ord(c) == 0xdc00:
|
|
|
|
# Hack to avoid combining this combining with the preceeding high
|
|
|
|
# surrogate, 0xdbff, when doing a repr.
|
2013-09-22 21:48:55 +00:00
|
|
|
c = '\\' + c
|
2013-09-22 20:39:16 +00:00
|
|
|
elif ord(c) in (0x2d, 0x5b, 0x5c, 0x5d):
|
|
|
|
# Escape regex metachars.
|
2013-09-22 21:48:55 +00:00
|
|
|
c = '\\' + c
|
2013-09-22 20:39:16 +00:00
|
|
|
categories.setdefault(cat, []).append(c)
|
|
|
|
|
|
|
|
f = open(__file__, 'w')
|
|
|
|
f.write(header)
|
|
|
|
|
|
|
|
for cat in sorted(categories):
|
2013-09-22 21:48:55 +00:00
|
|
|
val = ''.join(categories[cat])
|
2013-09-22 20:39:16 +00:00
|
|
|
if cat == 'Cs':
|
|
|
|
# Jython can't handle isolated surrogates
|
|
|
|
f.write("""\
|
|
|
|
try:
|
|
|
|
Cs = eval(u_prefix + r"%s")
|
|
|
|
except UnicodeDecodeError:
|
|
|
|
Cs = '' # Jython can't handle isolated surrogates\n\n""" % repr(val).lstrip('u'))
|
|
|
|
else:
|
|
|
|
f.write('%s = %r\n\n' % (cat, val))
|
|
|
|
f.write('cats = %r\n\n' % sorted(categories.keys()))
|
|
|
|
f.write('# Generated from unidata %s\n\n' % (unicodedata.unidata_version,))
|
|
|
|
|
|
|
|
f.write(footer)
|
|
|
|
f.close()
|